YouTube Transcript:
Promoting Competition in AI

Skip watching entire videos - get the full transcript, search for keywords, and copy with one click.

Video Transcript

good morning and welcome to Stanford for

our conference on promoting competition

in artificial intelligence my name is

Maria France I'm the director of the

business government and Society

initiative at the Stanford Graduate

School of Business artificial

intelligence or AI is the topic dour

from the classroom to the boardroom but

what makes today's conference unique is

its multifaceted look at the competitive

Dynamics across the AI landscape from

the chip to the app I'd like to thank my

colleagues at the Stanford Institute for

economic policy research and the

antitrust division of the US Department

of Justice for creating a compelling

program for many of us to learn and help

to shape future policies and strategies

so let's get started it's my privilege

to introduce our first Speaker Jonathan

caner Jonathan caner is the Assistant

Attorney General for the antitrust

division of the United States Department

of Justice he has served in that

position since November of 2021

throughout his career a ag caner has

been a leading advocate for strong and

meaningful antitrust enforcement and

competition policy having served as a

partner at two national leading law firms

firms

and as a founder of his own Boutique

antitrust Law Firm he began his career

as an attorney at the Federal Trade

commission's Bureau of competition

please join me in welcoming Jonathan [Applause]

caner well Maria thank you for that

incredibly warm introduction and for

having us here today um it's pretty

amazing um in academic

conference uh that is full first thing

in the morning and so um it really is I

think a statement to why this issue

matters so much and matters to so many

people there's a degree of urgency and

seriousness um and today's discussion is

really about bringing together people

who sometimes may not ordinarily come

together uh in the same room or on the

same uh panel and talk about issues that

are important and sometimes difficult to

talk about but

um that's what we're here to do it's

sometimes to confront the hard problems

uh and it's only when we do that that we

can actually start developing Solutions

um on beh half of the antitrust Division

I I want to thank Stanford Institute for

economic policy research uh and GSB uh

for hosting this conference it really is

amazing um but I would like to take a

moment uh if you will indulge me um to

acknowledge a particular specific person

here um our Chief Economist Susan ay um

who is not just the chief Economist of

the antitrust division but uh Stanford

Professor um she is one of a kind um

truly one of the smartest most

accomplished Minds in the world and has

been on the Forefront of so many of the

critical issues that um we're going to

talk about today uh and was doing that

before anyone else um I've had the

privilege and um and um honor of of

knowing Susan now for oh it must be 15

years and we've become good friends and

colleagues and thought partners and I'm

learning from Susan every day she is um

generous with her mind she is a leader

um a thought leader a leader of people

and somebody who um really is uh among

the smartest most talented and skilled

um thinkers of her generation in the

world and and we are honored every day

to work with her as I often say who

needs artificial intelligence when you

have the real thing and the real thing is

Susan um we meet today at the dawn of a

new technological Revolution AI has so

much promise but as with prior technological

technological

revolutions there will be threats to

confront um they're going uh ongoing

debates and we're hearing them all the

time and we'll hear them today uh that

impact Safety and Security um and talk

about how new technological developments

including AI can benefit Society but

also create risks for society uh the

antitrust laws play an important role uh

in this um development of new

technologies um there are um and it

exist broadly the antitrust laws to pre

preserve healthy and competitive markets

um today in this Workshop we're going to

focus on how to realize the competitive

promise of AI and reduce its threat to

competition in the American economy uh

and and reduce threats to competition in

the American economy um history teaches

us that effective antitrust enforcement

often coincides with major industrial

and technological change throughout

history we see examples of important

antitrust cases in fact some of the most

significant anti trust cases in history

that opened up markets and paved the way

for new companies and new Innovations to

arrive and to thrive um tr's Standard

Oil case is a famous early example known

to many splitting that oil trust into at

the time 34 independent companies today

we still um know and talk about and

purchase from The Offspring of Standard

Oil in companies including Exxon Mobile

and Chevron marathon and even BP which

merged with Amo and Arco um descends

from Standard Oil antitrust enforcement

actually resulted in the creation of new

competitors who thrived because of their

rivalry and we've seen lots of benefits

to this to society um in their

intervening Century we see a similar

story with uh the Bell System breakup

When AT&T was broken up into multiple

companies competition and telephone

markets thrived alongside those new

firms um today's a today we have AT&T

Verizon Lumen perhaps not as many as

some wish we would have uh but they are

direct descendants of successful

antitrust enforcement uh at the same

time that opened the technological space

to competition and created the room for

us Telecom and internet Industries to

develop and Thrive and more recently uh

the division's victory in the 20 1 case

against Microsoft for illegal

monopolization opened up the modern

digital economy and paved the way for

the success of today's firms and Beyond

many of whom inhabit um this very part of

California over and over again we see

that antitrust enforcement at moments of

industrial Evolution has the opportunity

to Spur innovation in its wake opening

the door to competition in new

competitors allows for the develop of

different business models and new

economies but we also see structures and

Trends in AI that should give us

pause AI relies on massive amounts of

data and computing

power which can give already dominant

firms a substantial Advantage powerful

Network and feedback effects may enable

dominant firms to control these new

markets an existing power in the digital

economy May create a power powerful

incentive to control emerging

innovations that will not only impact

our economy but the health and

well-being of our society and free expression

expression

itself these Technologies hold unbounded

promise for Innovations and change that

were once the exclusive domain of

Science Fiction at the same time AI

inputs and outputs have unique

characteristics that present new threats

to the markets for human ideas and innovation

innovation

generative AI for example leverages the

creations of humans knowledge paintings writing

writing

ideas absent competition to compensate

creators for their Works adequately

compensate creators for their Works AI

companies could exploit monopsony power

on levels that we have never seen before

with devastating

consequences everyone concerned about

human progress should be concerned about

that what incentive will tomorrow's

writers creators journalists thinkers

and artists have if AI has the ability

to extract their Ingenuity without appropriate

appropriate

compensation financially they will have

only those incentives that competition

competition between Foundation models

acting in concert with the IP system

creates in the absence of competition we

may see the problems Market power on the internet

internet

has caused in journalism spread to other

critical content creation

markets the people who create and

produce these inputs must be properly

compensated this of course is critical

to the Creator Community which is not

just about entertainment but

entertainment is very important but

about free

expression which is the most impactful

and beautiful form of human Innovation and

and

Ingenuity but it's also about so much

more it's about the Physicians and

patients whose Health Data is fed into

massive AI models it's about the

creators and artists whose words

thoughts and creativity are being

captured and used and of course the

journalists news outlets who are vital

to our

democracy there's more good news in the

history of industrial Evolution however

the ni trust laws adapt to changing

Market realities the principles of

competition enforcement apply whether an

innovation is powered by steam by

transistors or by reorganizing human

thought through machine learning at the

antitrust division we are actively

examining the AI

ecosystem both through our policy work

at events just like this

one and through our enforcement of the

Sherman and Clayton XS monopolizing

Upstream markets for Creative works is

monopolization whether or not a large

language model is involved D combining

to set prices with rivals in concerted

action whether or not an algorithm

assist that collision and risks fixing

prices if firms in the AI ecosystem

violate the antitrust laws the antitrust

division will have something to say about

about

that of course adapting our enforcement

program requires a deep understanding of

emerging AI ecosystem in different

sectors across the economy I'm

enormously excited to learn and to dis

discuss those issues today with an

incredibly impressive group of experts

today is really an important step

forward in our effort to learn and grow

it's our also an effort and this is

really I think a a distinctive feature

of today's

discussion um about bringing

stakeholders together from different

parts uh of the Continuum different

parts of the stack as we heard before

people who write code develop Hardware

write words and everything in

between so with that it's my pleasure to

kick us off and I'd like to introduce

vice president Jova she is the vice

president of the European commission for

values and

transparency from 2014 to

2019 vice president served as European

commissioner for justice consumers and

gender equality before that she served

as the minister for regional development

of the Czech Republic in her current

role vice president Jova leads the

commission's work on values and

transparency upholding the rule of law

and ensuring the Integrity of the

democratic system her deep well of

experience also

includes uh issues in consumer

protection data protection privacy

rights and importantly media Freedom we

are honored thrilled to have her with us

today thank you for joining us as part

of your us visit please join me in

you ladies and gentlemen this is exactly

what the AI will never replace for us

this human beings meetings and warm

words thank you very much for giving me

the opportunity to to speak uh at this

uh very important conference uh on a

very topical uh matter because uh the

artificial intelligence uh it's not the

matter of the future it's something we

live through and with already today and

uh over several years of my involvement

in the creation of the rules for AI and

digital world uh in the in the EU um I

had that feeling that somebody very

important is missing in that and it's

the sphere of research and science which

should have stronger voice I always was

uh participating in the interplay

between politicians and

Technologies and really the the voice of

research was missing and I think that

it's now time that uh the voice of

researchers and scientists is heard uh

more and uh where is Jonathan he

left uh I wanted also to thank thank him

and the Department of Justice for almost

10 years of fantastic

cooperation uh we worked uh together on

better understanding of uh the need uh

to protect privacy of people uh I had a

lot of discussions with the Department

of of Justice about how to cooperate

better in the field of Criminal Justice

and recently and this is my special

thanks uh we have very intense talks

about about how to uh bring Justice more

Justice into the I would call it

miserable War which is up and running in

Ukraine because we discuss about how to

uh establish the tribunal to punish the

uh war of aggression and and genocide

and we are also discussing how to use

the uh frozen assets from Russia to uh

uh reconstruct uh Ukraine so many things

in the times like this we only don't

need reliable Partners or Allies we need

friends and EU feels strongly especially

in these days that we have friends in

the United States so thank you very much

for that but coming to our today's topic

uh thank you for bringing uh the

spotlight into a very important

intersection of AI and

competition that is an overwhelming

agreement AI is a technology that can

bring many revolutionary

changes the paraphrase a quote from the

recent opam Heimer movie this isn't a

new technology it's a new

world it is one of the most powerful

Technologies ever invented and AI has a

potential of transforming everything not

only a competition

policy today I would like to talk to you

about two challenges of a I and the eu's

approach to it first AI challenges to

competition to stay faithful to theme

chosen by our hosts second I would like

to zoom out on the broader challenges

for AI of AI for democracy because this

is what is on my table in the European

commission this is at the heart of my

visit in California as part of what I

call a democracy tour in Europe we are

working hard to find answers to these

challenges brought to by digitalization

and shape our own approach to them we

start from a simple idea technology is

to serve the humans not the other way

around We Are Not Mere data fields we

are not just data donors for tech

companies to harvest from and then make

the decisions for us or make money on

our thoughts and

fears this is why we have chosen a

European model to regulate Technologies

in order to make sure they respect human

rights and I want to tackle one

skepticism that I have heard from many

concerns in the United States had on

namely that regulation stifles

Innovation I heard a lot of that when we

were introducing the general data protection

protection

regulation uh this uh concern this fear

did not materialize and I am sure it

will be also the case of AI I am sure

that Innovations and regulations can

work hand in hand we design laws to

address risks for the people or to open

markets that have been sealed by those

who have become too big to compete

against this brings Trust of of the

consumers and Innovations through

competition predictability and and

sufficient legal

certainty and where there is trust and

healthy competition there is investing

from both public and private

sources but there are indeed challenges

to competition and I'm sure you will

speak a lot about that

today the debate is very timely because

I still believe we have a narrow window

of opportunity to set the rules that

would allow competition to thrive in the

world of AI

in both the EU and the us we are facing

very similar problems to deal with the

challenge of the future it is good to

understand the past so we know where we

are coming from I am here thinking of

the digital disruption in the early

2000s social media digital marketplaces

online payments this was all so

exciting and this was the age full of

pioneer ing innovators some started in

the garage many of them at universities

or while dropping out of them the

innovators could not foresee or grasp

grasp

sorry the scale of impact their

Revolution would

bring today with a revolution some of

those things sound familiar but I also see

see

differences first the revolution is Led

in large part by incumbents powerful

companies with stakes in many many

markets second our awareness virtually

no one has doubts that AI will affect

every aspect of our work and life this

is what we didn't know at the beginning

of the

digitalization at the moment when some

boys started to invent things in the

garage I will speak with some of them here

here

also at my visit of

California uh the young inv inventors

who didn't know that one day they will

have impact on the whole world I used to

call them digital Gods when I came first

time to California in

2016 and now I think that we came to

agreement that we need meaningful rules

and for them the European rules are

quite demanding but what I hear from the

tech companies uh it's good that in

Europe which is such an important market

for us we have the necessary legal

certainty so but coming back from the

early uh Century the century and garage

uh coming back to

presents so our awareness is here now we

know that AI will affect every aspect of

our work and Life Third the technology

itself Ai and large language models have

he huge entry requirements they need

unprecedented amount of data and storage

among others the more the better in the

market where you need an entire

ecosystem to thrive it is hard to

imagine a teenager with a vision

challenging Microsoft and open eye AI or

metas Lama 2 or Google's de mind and as

many of you in the crowd today are

competition experts I am sure you can

immediately see barriers to entry

everywhere so how to answer these

challenges first I cannot help it we

have to deploy full human intelligence

and events like this we have to figure

it out and I'm sure we will second by

working together and this is why I

appreciate a Meeting of Minds with

Governor Newsome Europe and us must

continue to work together closely on

these issues to achieve a greater impact

and third by thinking out of silos

competition policy has to work together

with digital regulation that sets guard

rails in Europe This Means for instance

the digital markets

act let me now move to the challenges of

AI for

democracy democracy is an open debate

where we can argue and disagree seek for

compromise and then argue again we are

now few days ahead of European elections

I always say that the elections and

electoral campaigns have to remain fair

play of real human beings real Visions

people speaking to each other not the

competition of who will be better in

using manipulative techniques the this

is why are am I'm traveling now from

state to state in the EU and asking our

partners how they are ready to counter

these new

threats we have seen how digital

revolution has pushed us further into

bubbles or digital rabbit holes we are

still learning about the impact of

social media on our children and their

mental health and I do believe this

could be the main challenge for us in

the years to come

we are seeing in seeing in Europe how

Russia but also China Iran and other

actors use digital means to spread

disinformation and foreign

interference and ladies and gentlemen

this is a security risk to Europe and to

the United States as well Russia is

fighting in Ukraine with bombs and with

disinformation all over the world also

here in the United

States with AI they are just gaining a

new powerful tool to deploy old

tactics why they are doing this because

the essence of democracy is trust trust

to one another trust in the Democratic

institutions or the media without trust democracy

democracy

crumbles now with AI we can hardly trust

what we see or hear anymore deep fakes

synthetic images or videos not to

mention texts are fast becoming

indis IND

writer never again this [Laughter]

[Laughter]

word I was training the whole morning this

this [Laughter]

word but he wanted to

that uh but what I wanted to say yeah

the human eye cannot distinguish between

the false and the truth this is what is

say I already mentioned that I travel

from state to state on my democracy tour

I want to understand how well you can

countries and the authorities experts

and also the whole society are prepared

to digital threats to democracy and

security among other things to the fair or

unfair uh competition in electoral

campaign and this also is topic which

brings me to California because big Tech

has a tremendous role to play as well

what is being done here has a big impact

in Europe and in the rest of the world I

cannot cannot resist I have to share

with you the story when I first realized

that and I was here I think it was 2016

or 17 we had a big crisis in Spain the

separatists uh uh organized referendum

uh in

Catalonia and here in Silicon Valley I

received a lot of questions what what is

Catalonia Catalonia

and I didn't understand why they asked

me and they asked me because they had

hundreds of calls on closing this or

that website time is up I am in the

middle uh but this is a good story I

will finish

it uh that was the moment when I

realized that the boys in Silicon Valley

will not re decide on how our

referendums will

result and this was also the moment

which started first uh uh gentleman

agreements with the digital uh companies

and then the

regulation give me two more

minutes Madame

Professor thank

shter um yes uh we proposed an AI Act a

first comprehensive legislation on AI

including generative

AI we will enter it will enter into

force in July it is not only trust

fostering but also good for Innovation

first because of scale it means one set

of rules across all 27 European member

states second it gives the providers of

AI systems a stable and predictable

operating environment and sir said the

AI act encourages companies to develop

new products

easier I've been 10 years in the

commission over eight years we were

discussing how to regulate the digital

space eight years of discussions and

then regulation came the Digital

Services act for AI we feel we don't

have that time we had to act much

quicker so that's why we have the AI Act

and we also have uh relevant

International agreements like G7

agreement on secure uh

AI uh in the AI act uh we uh took the

risk-based approach we require the

companies to test the risky product and

the incidence with Tik Tock light or

recent launch of the Google search

product shows that this is only good for

consumers but good for companies as well

ladies and gentlemen the debate like you

are having today is not only needed but

it is urgent I don't have all the

answers I think nobody has this is why

we need Collective efforts Collective

wisdom to ensure competitive and

Innovative Market preserve democracy and

ultimately our values we still have a

chance to shape the AI Revolution so it

can become a Force for a positive change

rather than another risk that brings us

closer to a reality captured in a

dystopian TV show Black Mirror thank you

for your attention I wish you a very fruitful

fruitful [Applause]

debate okay I'll take the

so thank you so much Vice President

durova we are so honored to have you

here and I think this is maybe the the

first time in the history of timing that

uh someone actually listened to the

warning uh timing sign so uh so thank

you we we we really appreciate your

remarks so in this next part I'm I'm

Professor Susan ay from Stanford GSB and

I'm currently Chief Economist at the any

trust division at the doj and we're to

do the table setting for the rest of the

day we're really fortunate to have two

of the leading lights in AI from from

computer science and so we'll be

connecting the technology and the the

science to the competition issues we're

talking about today and to start off

with we are so fortunate to have Percy

leang here who's a professor of computer

science and all of the BIOS extended

bios for our speakers are on the website

so as a star AI professor at Stanford

and one of the leading lights of

national language processing you were

perfectly positioned to lead us as we

transitioned into this latest evolution

of AI um and indeed you were ahead of

the game um we were actually working

together at the Stanford um Institute

for human centered artificial

intelligence a few years ago and you had

the foresight to propose and start the

center for research on Foundation models

in 2021 before the rest of uh the world

was quite so aware of what this was all

going to look like um can you talk about

your vision for the center and why you

founded it and and how it's going yeah

yeah thanks for having me Susan so 2021

basically prehistory um in in AI terms

um for many uh chat gbt um 2022 December

was the sort of the turning point for me

it was May 2020 that's when opening

released their gpt3 model and that model

was a a huge model an order of magnitude

larger than before and it could it

demonstrated to me that this technology

and this mindset of training larger and

larger models was going to have a huge

impact on on society and at that time I

went around talked to many folks at the

uh at Hai and and say hey we got to pay

attention to this um you know

fortunately many people uh uh were also

uh compassionate um and agreed so we've

founded the center for research on

Foundation models um some people thought

I was crazy but that's um but um that's

fine turned out you were right maybe

still crazy both can be true um and and

but it was surprising it uh Chad gbt I

think really um demonstrated that this

was uh something to pay attention to and

it did come faster than I thought um but

so so what is foundation models about so

the idea of a foundation model is that

um someone can take a large amount of

data at scale usually scraped from the

internet and throw a huge amount of comp

computation power at it and this one

recipe can produce a single model that

is capable of demonstrating a large

range of tasks um not just high-end

generation uh but but also um you know

ability to solve many tasks so for the

first time we have more of these

generalist uh agents that um people had

aspired to for decades I've been a AI

researcher for you know a few decades

and this moment is very uh special um so

it's somehow the the the idea that you

can take you know capital and put it in

and sort of predictably get U more and

more capabilities that that was a sort

of a mindset uh change and I want to

also stress that Foundation models isn't

just about chachal language models it's

a much more general idea that any time

in any domain where you have a huge

amount of data whether it be um you know

Vision Videos time series retail data um

computation biology astronomy all these

areas we're seeing being transformed

because of this kind of remarkably

simple idea where you have a Transformer

architecture plus um you know gradient

Ascent plus um you know gpus um allows

you to in any gain give you sort of

remarkable uh

properties great thank you so much and

so what do the the center actually want

to do what are what are your goals in

this and hopes and and how you can

contribute to the conversation about

them yeah so why do we create this

Center so one thing that we were

reacting to was that at that time since

2020 and you see a kind of progress um

more in this direction is that the level

of transparency and openness that we see

in AI has diminished right in the

2010s when the Deep learning um

Revolution was um you know going strong

um there was a culture of researchers um

sharing data uh code models openly and

various companies would be able to uh

startups would be able to leverage this

and I think a lot of the Innovation came

from um this this openness but what has

happened over the last three years is

kind of the as capabilities go up uh

openness uh goes down so from the idea

that um initially the model weights um

which is something hopefully we'll get

to a little bit later are no longer

available to you know the stop of uh

Publications um and gating of access and

and this I think from a academic

researcher perspective was quite

concerning and not only from a

researcher perspective where maybe

Academia could no longer engage with

this important technology but also from

a societal aspect if we have this hugely

important technology which we don't

understand and as is is locked away

behind you know a guarded closely then

uh we might um be in a bad situation so

the center was created to with a mission

of making Foundation model is more

accessible more transparent um for uh

the world and so what are the kind what

can you have some examples of what how

that transparency can be beneficial what

are some of the sort of use cases from

the center that that that would be

helpful for from that transparency yeah

so one of the key things that we're

focused quite heavily on is uh

benchmarking and evaluation so we have

these models that are being developed

and released and people are using them

but we have shockingly little rigorous

evidence of what their actual

capabilities and risks are there are a

bunch of demos and things are being

deployed and one thing that we've

undertaken at that Center is developing

benchmarks so we developed something

called the holistic evaluation of

language models uh two years ago which

we've been extending o over time to look

at all the different capabilities and

risks um um of these models as best as

we can and to try to uh Benchmark them

and also make all these results uh

publicly available for people to to see

and one of the things that's difficult

about benchmarking is that because these

models are so General um how do you even

you know test this I don't like usually

making analogies with humans but it it's

kind of like humans you're measuring the

cap abilities of a of a human being

because they they are so versatile and

so this is something that's very very

tricky but um I think we've made a lot

of progress on this uh covering many

different um tasks from question

answering to summarization now we're

looking at multimodal models for visual

question answering and image generation

and looking at um starting to look at

other dialogue uh settings so and I

think one thing that's important is

having a third party sort of neutral

third party that's doing these

evaluations and committing to doing them

across all the the foundation models

that are um are you know Frontier and

and uh reporting the results in a

standardized uh way yeah so is it is it

fair to say that as of a couple of years

ago we didn't have a lot of science

about how you would evaluate these types of

models yeah so I think um and we still

don't have a lot lot of science behind

how to evaluate these mod models but we

have a a little bit more now um I think

the the field of evaluation is just very

much underdeveloped I think in the

history of machine learning it used to

be the case that you define the

benchmarks and these benchmarks would be

very difficult for example imag net when

imag net was created no vision system

could really do anything respectable on

it and it was a sort of a North star for

many researchers to keep on working on

image and then so that that was a

database with a bunch of images and

labels and so you could just test out

how your model worked against it and

there was a number right A number that

said okay you beat the last Benchmark

I'm the star of the AI conference this

year yeah yeah and now I think it's sort

of flipped where people are developing

models and now benchmarks are trying to

catch up to capture what these models

are doing and on top of that we're no

longer interested in just the accuracy

but we're interested in things like bias

and fairness and robustness and whether

the models uh know what they don't know

and not to mention different types of

risks uh you know we heard about

disinformation but we heard about you

know their cyber risks and all these

different risks and how do you measure

them in a way that can inform you know

uh businesses and and policy makers and

so do you think it's possible to do that

science within like one research group

in one place or um do you think you you

know need a larger Community working on

those problems I I think you definitely

need a larg lger community and this is

presum uh the way that Academia works is

that it's it's the community we do a lot

of work but there's also good work

happening at various places like you

know University of Washington and um

many other uh universities where these

data sets are created and there's often

um benchmarks that are created that

derive from these data sets and then

they're put together in you know in

different ways um and so so a lot of the

you know meth what we're missing is sort of

methodological um you know science for

how to conduct benchmarks I mean just as

a simple example one of the things that

you know chat gbd can do is it can

generate you know an essay or a summary

and this alone is is a very deep you

know problem how do you evaluate um

something that's so complex and of

course you run into cases where if it's

on a topic that is is very Niche then

hiring human experts is is going to be

very expensive so can you use uh

language models themselves to evaluate

but then you don't trust the language

models so it's kind of it's very

interesting I think technically to see

what you can do and I think you can

actually make some progress even though

it seems like you're kind of

bootstrapping yourself um great well so

so but in general so it sounds like your

Center is promoting the ability ility

for lots of researchers to be able to

contribute to this relatively nent but

important field um so do you could you

talk us through other than the open

weights what do you actually need to to

be able to do to do that kind of work so

if we want to have a large community of

people contributing to this nent science

which might take a long time to figure

out because as you mentioned the will

probably still be developing new

evaluation metrics in 10 years for new

risks that have been identified what

what do what do people need to be able

to contribute to that whether it's

researchers or whether it's startups

that are creating um tools that can help

other businesses deploy them safely like

what are the inputs to make that to

contribute to that yeah I think one

thing I'll start by saying is um in

Academia you know we're meant to do

fundamental research that moves the

needle and a lot of what we are missing

right now is just you know computer

um I think we can do a lot by taking

existing models benchmarking them um we

can even if we have open weights for in

the case of llama three or llama 2 we

can inspect them but I think there's no

substitute for actually being able to

train um you know models um for example

we're developing different types of

model architectures which are uh

different from the

Transformer um and to demonstrate the

the the value of those we need to be

able to train and the the resource

differential between Academia industry

is just multiple orders of magnitude and

I think it's unrealistic to expect that

them to be equal but I think we as a um

you know uh Community um let's say we

need to invest more in in computation so

that we can at least be in um playing

playing the game so just to make sure I

understand so there's a couple of

different kinds of research you can do

one is that you can take either a closed

or an open model if for the closed you

have to pay for it and then you can

evaluate it using evaluation metrics but

then there's other research that you can

do that kind of gets into the guts of it

that you could only do by training your

own models perhaps at smaller scale and

so could you talk about what's needed

for the different types and what other

inputs you might need for the second one

yeah so for the first typee which is

taking existing models and studying them

um you need access for and for the most

part we do have access um I think it

would help be helpful to have deeper

access for example with uh gp4 you can

only get access to Generations you can't

look inside what the model activations

uh um and there's also we don't have

access to training data so basic

questions in machine learning such as

did you is your test set in your

training set we have no ability to

really know for sure so there's some

asterisks on um the ability to do you

know proper research there but I would

say it's in a much better State and then

I would say this latter category of

models uh research which involves

intervening on the the architecture or

you know data um those require

substantially more resources and I think

um the the and I I think that you need

enough resources to be able to train a

large enough model to show that um it

can do something right so one of the

phenomenon that you see with um these

models is this emerging Behavior you

need to be at a certain scale before you

start seeing anything happening

otherwise it looks like nothing is

working nothing is is is

working and and I think the reason why

this that category is important but also

under resource is that you know we talk

about all these problems with these

models with AI biases uh you know

hallucinations people can you know

jailbreak and get these models to behave

in sort of unpredictable ways I think

these are really fundamental problems

that these models have and we're not

going to be able to solve them just by

you know patching things away um and

this requ requires really addressing the

the fundamental core of like how the data interacts with the model and

data interacts with the model and understanding that in a much um deeper

understanding that in a much um deeper way which unfortunately I think we don't

way which unfortunately I think we don't have the community doesn't have as much

have the community doesn't have as much patience or for this um but but I'm sort

patience or for this um but but I'm sort of pushing for more resourcing so we can

of pushing for more resourcing so we can actually get to the bottom of things and

actually get to the bottom of things and and this is a long-term agenda to get to

and this is a long-term agenda to get to the bottom things and hopefully develop

the bottom things and hopefully develop a a stronger

a a stronger Foundation great well so um one of the

Foundation great well so um one of the things you know when when you're doing

things you know when when you're doing the kind of uh fine-tuning or retrieval

the kind of uh fine-tuning or retrieval augment and generation or other types of

augment and generation or other types of uses of of open models you even

uses of of open models you even researchers or small startups need

researchers or small startups need resources they need compute it's it's

resources they need compute it's it's something that's been kind of amazing to

something that's been kind of amazing to me that um it's sort of like the first

me that um it's sort of like the first time in my lifetime where you can put

time in my lifetime where you can put out a problem and you know Stanford

out a problem and you know Stanford Master students can come out with

Master students can come out with something working in like 10 days which

something working in like 10 days which is it seems like kind of a miracle

is it seems like kind of a miracle because everything you think is going to

because everything you think is going to take two months and it takes two years

take two months and it takes two years but here people are are getting like

but here people are are getting like working prototypes up very very quickly

working prototypes up very very quickly all over campus and across the startup

all over campus and across the startup land and there's there's a number of of

land and there's there's a number of of pieces that go into making that possible

pieces that go into making that possible um so could you talk through um what

um so could you talk through um what some of those pieces are for

some of those pieces are for applications and and in particular I

applications and and in particular I think you co recently co-founded a

think you co recently co-founded a startup called together. that kind of

startup called together. that kind of contribut

contribut some of the pieces people might use but

some of the pieces people might use but I'd love to hear about about together

I'd love to hear about about together and what it does and also what are what

and what it does and also what are what are the pieces that go into this that

are the pieces that go into this that can make this spur Innovation you know

can make this spur Innovation you know across all of everyone yeah yeah I mean

across all of everyone yeah yeah I mean to start out with I mean I completely

to start out with I mean I completely agree with you it's sort of amazing what

agree with you it's sort of amazing what one can do even in the afternoon with

one can do even in the afternoon with these these models um because

these these models um because the because I think at some level if you

the because I think at some level if you can write English or or in maybe another

can write English or or in maybe another English you can write a prompt and you

English you can write a prompt and you can get you can sort of just State what

can get you can sort of just State what you want this model to do and if you're

you want this model to do and if you're lucky enough then it can actually do it

lucky enough then it can actually do it so as a result um this technology has

so as a result um this technology has really opened up the opportunity for

really opened up the opportunity for many people who are not uh don't have

many people who are not uh don't have machine learning phds to be able to

machine learning phds to be able to interact which I think is wonderful and

interact which I think is wonderful and we're seeing a lot of um you know uptake

we're seeing a lot of um you know uptake in other areas like law and and Medicine

in other areas like law and and Medicine where this normally wouldn't have um

where this normally wouldn't have um happened um I will uh put a little bit

happened um I will uh put a little bit asteris is on the the fact that there's

asteris is on the the fact that there's a gap between guine a demo up where you

a gap between guine a demo up where you look at it it's like wow that's pretty

look at it it's like wow that's pretty cool to an actual reliable product that

cool to an actual reliable product that you can ship to many many people so that

you can ship to many many people so that last mile problem is still um

last mile problem is still um challenging it might be a little bit

challenging it might be a little bit easier but um don't get too excited and

easier but um don't get too excited and I think let me just pause you there that

I think let me just pause you there that is a really excellent point and that

is a really excellent point and that I've certainly heard that from a lot of

I've certainly heard that from a lot of of people across the industry as well

of people across the industry as well because if you're going to if if you

because if you're going to if if you want to put something in front of your

want to put something in front of your end customers it's not enough that it

end customers it's not enough that it mostly is right it has to not say

mostly is right it has to not say offensive things or do something that

offensive things or do something that damages your reputation or creates risk

damages your reputation or creates risk and so that's partly where your

and so that's partly where your evaluation stuff comes in that those

evaluation stuff comes in that those tools are an important part but so let's

tools are an important part but so let's just talk though for the moment about

just talk though for the moment about getting to the demo step even like what

getting to the demo step even like what what um what are the pieces the inputs

what um what are the pieces the inputs to that so from an application developer

to that so from an application developer perspective right so you have your

perspective right so you have your problem um the thing that is probably

problem um the thing that is probably the easiest to start is you take gp4 or

the easiest to start is you take gp4 or Claud or gemini or if your favorite

Claud or gemini or if your favorite model and you prompt it just to see get

model and you prompt it just to see get a sense of what it can do right and this

a sense of what it can do right and this is I would really call this the

is I would really call this the prototyping phase right you're ideating

prototyping phase right you're ideating and you can brainstorm while you're

and you can brainstorm while you're getting feedback from this actual system

getting feedback from this actual system which is I think sort of incredible kind

which is I think sort of incredible kind of user interaction um uh new type of us

of user interaction um uh new type of us user interaction um and then I think

user interaction um and then I think there's a but of course that's not

there's a but of course that's not enough and it's also expensive to use

enough and it's also expensive to use keep on using gp4 so one thing that's I

keep on using gp4 so one thing that's I think also important is that if you're

think also important is that if you're you know you're a business and you have

you know you're a business and you have uh you know your own data and your own

uh you know your own data and your own particular needs and businesses you have

particular needs and businesses you have to have Generations that have a certain

to have Generations that have a certain tone or style or you don't want to have

tone or style or you don't want to have it say XYZ then this is where

it say XYZ then this is where fine-tuning um you know comes in you um

fine-tuning um you know comes in you um have to be able to you know spec fine

have to be able to you know spec fine tuning basically says I have I'm more

tuning basically says I have I'm more opinionated about what my system wants

opinionated about what my system wants to do so I'm going to take some work to

to do so I'm going to take some work to really flesh out here the cases where it

really flesh out here the cases where it should do X and here's the cases where

should do X and here's the cases where it should do y and then you you

it should do y and then you you fine-tune a model and on on top of that

fine-tune a model and on on top of that um you can connect these language models

um you can connect these language models with uh different types of of tools so

with uh different types of of tools so interacting with the the the world and

interacting with the the the world and this is you mentioned rag which stands

this is you mentioned rag which stands for retrieval augmented generation the

for retrieval augmented generation the idea that you can take your do internal

idea that you can take your do internal documents um and be able to look at them

documents um and be able to look at them and have that influence your generation

and have that influence your generation or you can if you have a database or

or you can if you have a database or different apis inside uh your your

different apis inside uh your your organization you can um query those as

organization you can um query those as well and have the language model

well and have the language model essentially as a um you know a sort of a

essentially as a um you know a sort of a driver that c dispatches to different

driver that c dispatches to different services and Aggregates and then

services and Aggregates and then generates the result May summarize it at

generates the result May summarize it at the end so if you want to build

the end so if you want to build something like that what do you what

something like that what do you what what do you need to build it and and how

what do you need to build it and and how does together. fit in yeah so together

does together. fit in yeah so together was founded two years ago on the premise

was founded two years ago on the premise that we wanted to build a platform that

that we wanted to build a platform that enabled um you know developers to be

enabled um you know developers to be able to use this technology in a easy

able to use this technology in a easy and cost effective and you know

and cost effective and you know performant way as as possible um so one

performant way as as possible um so one of the services that we provide is

of the services that we provide is simply uh know inference if you have

simply uh know inference if you have let's say a llama 3 an open model that

let's say a llama 3 an open model that you want to prompt and run you can use

you want to prompt and run you can use together just like the way you would use

together just like the way you would use a GPD 4 but at a you know fraction of

a GPD 4 but at a you know fraction of the price um you can also find it's kind

the price um you can also find it's kind of like a cloud computing solution for

of like a cloud computing solution for this that could host the model yeah so

this that could host the model yeah so so I I think one yeah that's a good

so I I think one yeah that's a good point so one thing that um people might

point so one thing that um people might maybe not quite realize is that there is

maybe not quite realize is that there is the open Ai and in Googles of the world

the open Ai and in Googles of the world that have their ape services and it's

that have their ape services and it's very convenient and then on the other

very convenient and then on the other hand you might hear these open weight

hand you might hear these open weight models like llama 3 or or mistol or

models like llama 3 or or mistol or mixol where you get the weights which is

mixol where you get the weights which is great for customization but you might

great for customization but you might hear that oh okay well I have to host it

hear that oh okay well I have to host it myself and I have to deal with it but

myself and I have to deal with it but that's not not really true because

that's not not really true because there's actually um multiple services

there's actually um multiple services including together that host these for

including together that host these for you so you really have the convenience

you so you really have the convenience of an an open AI like uh framework but

of an an open AI like uh framework but um you know cheaper and you get the

um you know cheaper and you get the benefits of transparency so you might

benefits of transparency so you might you might you load up your data so like

you might you load up your data so like when I see students doing here they

when I see students doing here they might load up their data they then could

might load up their data they then could fine-tune one of those open models using

fine-tune one of those open models using the the compute and and sort of pay as

the the compute and and sort of pay as you go but then there are also software

you go but then there are also software tools that people are using as well so

tools that people are using as well so the students don't necessarily have to

the students don't necessarily have to write all the code from scratch to do

write all the code from scratch to do the fine tuning can you just say briefly

the fine tuning can you just say briefly like how that step works so if you want

like how that step works so if you want to fine tuned there's no code actually

to fine tuned there's no code actually um you upload a file that has examples

um you upload a file that has examples of input output behaviors this is an

of input output behaviors this is an instruction um and this is the response

instruction um and this is the response you put them in a file you upload it you

you put them in a file you upload it you click you know finetune and then you get

click you know finetune and then you get your fine-tune model and then you can do

your fine-tune model and then you can do inference on this just like you would be

inference on this just like you would be able to do inference on any other um

able to do inference on any other um non-custom model yeah and so if we'

non-custom model yeah and so if we' wanted to do something like that like 15

wanted to do something like that like 15 years ago you know this would have been

years ago you know this would have been like you know all the time writing the

like you know all the time writing the code for the sastic gradient descent for

code for the sastic gradient descent for the the debugging and everything else

the the debugging and everything else but basically now literally you don't

but basically now literally you don't you you you can produce your own version

you you you can produce your own version with your own documents by pushing a

with your own documents by pushing a button yeah yeah and I think it's even

button yeah yeah and I think it's even better than that because 15 years ago

better than that because 15 years ago there were no Foundation models so you

there were no Foundation models so you basically have to start from scratch you

basically have to start from scratch you would gather your data set and then the

would gather your data set and then the problem is if you're starting from

problem is if you're starting from scratch you need a fair amount of data

scratch you need a fair amount of data you might need you know thousands or

you might need you know thousands or tens of thousands of examples to even

tens of thousands of examples to even get something that

get something that was interesting to play with and that

was interesting to play with and that meant you needed to First go you know

meant you needed to First go you know get a annotation team like hire

get a annotation team like hire annotators to create create this data

annotators to create create this data and you spend you know potentially

and you spend you know potentially months training animators getting the

months training animators getting the data and then you can train your model

data and then you can train your model and then you can see what it does and

and then you can see what it does and then if it doesn't work then you have to

then if it doesn't work then you have to go back but now because of foundation

go back but now because of foundation models you don't need that much data

models you don't need that much data because much of the common things about

because much of the common things about how you know English works or um you

how you know English works or um you know simple reasoning patterns the model

know simple reasoning patterns the model already knows about so all you need to

already knows about so all you need to do is supply information about your

do is supply information about your specific needs

specific needs and so you might be able to get away

and so you might be able to get away with you know hundreds of examples so so

with you know hundreds of examples so so the iteration Cycles are much shorter

the iteration Cycles are much shorter and then when you do prompting it's

and then when you do prompting it's often five examples or zero examples so

often five examples or zero examples so so I think that iteration cycle is much

so I think that iteration cycle is much much shorter great so actually we're

much shorter great so actually we're just running short on time but I'm going

just running short on time but I'm going to try to squeeze in my last favorite

to try to squeeze in my last favorite question um we have a lot of debates

question um we have a lot of debates about the risks and benefits of open and

about the risks and benefits of open and closed as an economist I'm particularly

closed as an economist I'm particularly interested in how you think about

interested in how you think about incremental risk of these open weight

incremental risk of these open weight models and could you just say briefly

models and could you just say briefly how you think about that yeah so one of

how you think about that yeah so one of the things that you might hear is that

the things that you might hear is that open weight and I use the word open

open weight and I use the word open weight instead of Open Source

weight instead of Open Source deliberately to be more specific is

deliberately to be more specific is about these model is that well people

about these model is that well people can use them for for harm you can f

can use them for for harm you can f tunea you can turn off the safety and

tunea you can turn off the safety and you can spread disinformation and and so

you can spread disinformation and and so on and I think it's really important to

on and I think it's really important to frame this conversation in terms of the

frame this conversation in terms of the whole picture which is first of all that

whole picture which is first of all that this model is a tool that is is uh used

this model is a tool that is is uh used um to potentially generate some uh some

um to potentially generate some uh some you know harmful content but then this

you know harmful content but then this harmful content also needs to be

harmful content also needs to be disseminated or in the case of let's say

disseminated or in the case of let's say biorisk you need to also you know

biorisk you need to also you know synthesize the you know the proteins and

synthesize the you know the proteins and so you need to think about the whole you

so you need to think about the whole you know picture and where you want to you

know picture and where you want to you know gate um another thing is that

know gate um another thing is that um the just because a model generates

um the just because a model generates let's say you know how do I make a bomb

let's say you know how do I make a bomb okay here's how you make a bomb um

okay here's how you make a bomb um doesn't mean that we should immediately

doesn't mean that we should immediately just shut it off because you can also

just shut it off because you can also you know find things like this on the

you know find things like this on the internet um you can find a lot of things

internet um you can find a lot of things about um you know um how to make

about um you know um how to make Chemical or bioweapons on you know on

Chemical or bioweapons on you know on Wikipedia and so there's a lot of other

Wikipedia and so there's a lot of other uh things so one thing that we've been

uh things so one thing that we've been trying to get people to uh think about

trying to get people to uh think about is the marginal risk of a particular

is the marginal risk of a particular technology with respect to what exists

technology with respect to what exists and and and I think only then can you

and and and I think only then can you make a meaningful calculation of you

make a meaningful calculation of you know whether introducing this um is

know whether introducing this um is going to what is the risk there um

going to what is the risk there um compared to the marginal benefits of

compared to the marginal benefits of having um these models and I I think the

having um these models and I I think the you know right now I don't see you know

you know right now I don't see you know substantial evidence that the marginal

substantial evidence that the marginal risk is is high and open AI actually

risk is is high and open AI actually also did you know certain studies the

also did you know certain studies the current models just aren't that um

current models just aren't that um impressive when it terms comes to like

impressive when it terms comes to like enabling much a broader uh surface area

enabling much a broader uh surface area of tax but this goes back to the point

of tax but this goes back to the point about evaluation we need to much more

about evaluation we need to much more rigorously evaluate the risks before we

rigorously evaluate the risks before we jump to conclusions and get our our

jump to conclusions and get our our hells and speculate about

hells and speculate about what you know could potentially happen

what you know could potentially happen well you've made every um intro

well you've made every um intro economics professor's heart sing with

economics professor's heart sing with marginal benefits and marginal cost so I

marginal benefits and marginal cost so I think that's a great place to close

think that's a great place to close thank you so much for being here thank

thank you so much for being here thank you great thank

you so next I'd like to welcome to the stage Andrew ing please come on up

stage Andrew ing please come on up Andrew well um and I just to remind

Andrew well um and I just to remind everyone if you're just tuning in all of

everyone if you're just tuning in all of the speaker bios are online um and so

the speaker bios are online um and so we're will'll be short with

we're will'll be short with introductions Andrew I think just to get

introductions Andrew I think just to get started you have a few slides to help

started you have a few slides to help frame some of your thoughts about open

frame some of your thoughts about open models open source and the role of

models open source and the role of competition and AI so I would love it if

competition and AI so I would love it if you could just kick us off with uh with

you could just kick us off with uh with with a few comments awesome thank thank

with a few comments awesome thank thank you Susan um it's good to hear pressy

you Susan um it's good to hear pressy good to see everyone so um I just Rec

good to see everyone so um I just Rec siiz and I'll be quick you know I I

siiz and I'll be quick you know I I think one of the hard things to

think one of the hard things to understand about AI is that is a general

understand about AI is that is a general purpose technology meaning is not useful

purpose technology meaning is not useful just for a single thing but it's useful

just for a single thing but it's useful for a very large multitude of tasks um I

for a very large multitude of tasks um I think there are some business

think there are some business representatives in the audience I think

representatives in the audience I think for businesses this is exciting because

for businesses this is exciting because it gives us the opportunity to identify

it gives us the opportunity to identify and build tons of applications in

and build tons of applications in healthcare Financial Services education

healthcare Financial Services education Logistics and on and on and on um and in

Logistics and on and on and on um and in terms of the registry landscape I think

terms of the registry landscape I think my my my key messages are one I think

my my my key messages are one I think it's important to distinguish between

it's important to distinguish between technology and applications um and

technology and applications um and second to promote competition and

second to promote competition and Innovation I feel like we should lean

Innovation I feel like we should lean into regulating applications rather than

into regulating applications rather than Technologies and and also um really

Technologies and and also um really promotes open source so um you know this

promotes open source so um you know this is what I think of as an AI stack

is what I think of as an AI stack different people have slightly different

different people have slightly different versions of this but I think the lowest

versions of this but I think the lowest level is a semiconductor level on top of

level is a semiconductor level on top of that are the clouds both of these are

that are the clouds both of these are relatively Capital intensive thus maybe

relatively Capital intensive thus maybe you know slightly concentrated and then

you know slightly concentrated and then the new layer has been the AI technology

the new layer has been the AI technology tooling layer the these provide

tooling layer the these provide Foundation models um um en large

Foundation models um um en large language models and so on and what I

language models and so on and what I seen with most waves of um Innovation is

seen with most waves of um Innovation is the media traditional media and social

the media traditional media and social media tends to focus on technology layer

media tends to focus on technology layer this is fun to talk about it's fun to

this is fun to talk about it's fun to talk about what's opening eye doing you

talk about what's opening eye doing you know and so on but I think that um

know and so on but I think that um there's an even bigger set of

there's an even bigger set of opportunities than this which is the

opportunities than this which is the application layer because for the

application layer because for the technology tooling layer and the layers

technology tooling layer and the layers below to be successful almost by

below to be successful almost by definition we need to applications on

definition we need to applications on top of them to generate even more

top of them to generate even more Revenue so they can afford to pay the

Revenue so they can afford to pay the technology layers underneath um and I

technology layers underneath um and I think in order to prevent choke points I

think in order to prevent choke points I think there's a lot of concern about

think there's a lot of concern about choke points AI technology layer we only

choke points AI technology layer we only a small number of companies have access

a small number of companies have access to Foundation models I think open source

to Foundation models I think open source is one of our best tools for promoting

is one of our best tools for promoting Innovation and preventing a new CH point

Innovation and preventing a new CH point from arising at AI technology layer and

from arising at AI technology layer and candidly I've been I would love if um

candidly I've been I would love if um governments all around the world can

governments all around the world can promote competition candidly I'd be

promote competition candidly I'd be quite happy if governments all around

quite happy if governments all around the world just don't stifle competition

the world just don't stifle competition but having having governments promote

but having having governments promote competition as opposed to merely avoid

competition as opposed to merely avoid stifling it seems like a a dream at this

stifling it seems like a a dream at this point in time but let me dig into that a

point in time but let me dig into that a little bit more um and and this is what

little bit more um and and this is what I mean by um technology versus

I mean by um technology versus applications and this is an important

applications and this is an important concept for all in government I hope I

concept for all in government I hope I hope you remember this part so I'm going

hope you remember this part so I'm going to talk about technology later

to talk about technology later as was an application layer so um an

as was an application layer so um an example of a technology might be an

example of a technology might be an electric motor and an electric motor

electric motor and an electric motor which can be used in lots of

which can be used in lots of applications could be put into a blender

applications could be put into a blender electrical vehicle a dial machine um or

electrical vehicle a dial machine um or also a guided bomb um and so I think of

also a guided bomb um and so I think of Technology as tools that can be applied

Technology as tools that can be applied in many different ways to solve many

in many different ways to solve many different problems like electric motor

different problems like electric motor um and then applications are specific

um and then applications are specific implementations designed to meet a

implementations designed to meet a particular customer need and this could

particular customer need and this could be a very beneficial need like a

be a very beneficial need like a dialysis machine or one that we don't

dialysis machine or one that we don't want adversaries to have access to like

want adversaries to have access to like a guided bomb um and I feel like if you

a guided bomb um and I feel like if you ask me to make electric motors safe I

ask me to make electric motors safe I don't know how to do that right the only

don't know how to do that right the only way to make an electric motor save is to

way to make an electric motor save is to make it so tiny that's basically useless

make it so tiny that's basically useless to everyone but and indeed blenders can

to everyone but and indeed blenders can be dangerous this is why we regulate

be dangerous this is why we regulate blenders we don't want blenders to you

blenders we don't want blenders to you know be dangerous in people's homes and

know be dangerous in people's homes and and damage and hurt children's Limbs and

and damage and hurt children's Limbs and so on um electric vehicles let's make

so on um electric vehicles let's make sure electric vehicles are safe and D

sure electric vehicles are safe and D machines boy that's a medical device

machines boy that's a medical device let's make sure to regulate that and

let's make sure to regulate that and guide the bombs yes we absolutely need

guide the bombs yes we absolutely need to regulate production of weapons and

to regulate production of weapons and Munitions um but if you try to say an

Munitions um but if you try to say an electric motor could be used by build a

electric motor could be used by build a bomb so let's make the electric motor

bomb so let's make the electric motor manufacturer guaranteed their electric

manufacturer guaranteed their electric motor will never be used by anyone to

motor will never be used by anyone to make a guide the bomb then the only way

make a guide the bomb then the only way is for the electric motor manufacturer

is for the electric motor manufacturer to shut down the business and that means

to shut down the business and that means you lose the blender EV and the DAT

you lose the blender EV and the DAT machine as well and this is what we're

machine as well and this is what we're seeing in in ai ai models lar language

seeing in in ai ai models lar language models Foundation models are general

models Foundation models are general purpose Technologies uh we can be use to

purpose Technologies uh we can be use to build a medical device to give medical

build a medical device to give medical advice um or to rank someone's Social

advice um or to rank someone's Social Media fee for good purposes or bad

Media fee for good purposes or bad purposes um it can be used to build a

purposes um it can be used to build a chat bot um it can be used to generate

chat bot um it can be used to generate political defects there are tons of

political defects there are tons of beneficial as well as a small number of

beneficial as well as a small number of harmful applications and I think risks

harmful applications and I think risks are a function of of the application not

are a function of of the application not to technology um and when you reg if

to technology um and when you reg if someone tries to regulate the general

someone tries to regulate the general purpose technology like electric motors

purpose technology like electric motors you know that tends to just stifle

you know that tends to just stifle Innovation because you're cutting off

Innovation because you're cutting off the Innovation at the root and shutting

the Innovation at the root and shutting down a huge range of beneficial

down a huge range of beneficial applications as well as maybe a small

applications as well as maybe a small number of harmful ones um so I feel like

number of harmful ones um so I feel like what we do need is um I do want medical

what we do need is um I do want medical devices to be safe I do want electrical

devices to be safe I do want electrical vehicles to be safe and what I'm seeing

vehicles to be safe and what I'm seeing unfortunately is many governments are

unfortunately is many governments are saying boy it's too hard to look at the

saying boy it's too hard to look at the applications that regulate them so let's

applications that regulate them so let's do the easy thing which is you know

do the easy thing which is you know let's just regulate the AI technology

let's just regulate the AI technology but that's a bad move I think we need to

but that's a bad move I think we need to do the harder but the right thing which

do the harder but the right thing which is look at the actual applications which

is look at the actual applications which where the risk are and pass appropriate

where the risk are and pass appropriate regulations so to to to protect um to

regulations so to to to protect um to protect citizens um and I've been really

protect citizens um and I've been really surprised at the intensity of the

surprised at the intensity of the lobbying um to shut down open source I

lobbying um to shut down open source I think maybe predictable there are some

think maybe predictable there are some you know companies that have invested

you know companies that have invested billions of dollars into training large

billions of dollars into training large Foundation models and when you've

Foundation models and when you've invested billions of dollars is really

invested billions of dollars is really inconvenient when some other company

inconvenient when some other company releases an open SCE or open weights

releases an open SCE or open weights model that severely degrades the value

model that severely degrades the value of that massive investment you made to

of that massive investment you made to train propriety model and I get it it's

train propriety model and I get it it's really annoying when someone open

really annoying when someone open sources something um but what has

sources something um but what has happened over the last year was really

happened over the last year was really intense lobbying under the guise of AI

intense lobbying under the guise of AI safety and you hear very smart people

safety and you hear very smart people say say things like of course I support

say say things like of course I support open source but don't you want open

open source but don't you want open source to be safe and and that's used as

source to be safe and and that's used as an excuse to put in place very

an excuse to put in place very burdensome licensing requirements that

burdensome licensing requirements that will stifle anyone's ability you know

will stifle anyone's ability you know that they'll severely stifle many team's

that they'll severely stifle many team's ability to release free open weights of

ability to release free open weights of Open Source software on the internet um

Open Source software on the internet um and you know in and and candidly I I

and you know in and and candidly I I think that um open source today open

think that um open source today open weight is open source is a key part of

weight is open source is a key part of the AI supply chain um I feel like the

the AI supply chain um I feel like the the one of the latest lines of attacks

the one of the latest lines of attacks on open source has been um to prevent

on open source has been um to prevent you know near peer adversaries from from

you know near peer adversaries from from getting access to it but candidly I I

getting access to it but candidly I I actually on on that ground is actually

actually on on that ground is actually worry I think um uh you know someday

worry I think um uh you know someday right some kid in some developing

right some kid in some developing economy will going to large out of

economy will going to large out of foundation model also large language

foundation model also large language model you know is democracy important

model you know is democracy important what's the role of Free Press is

what's the role of Free Press is independent Judiciary important and when

independent Judiciary important and when kids like that ask questions I would

kids like that ask questions I would quite like the lar language model to

quite like the lar language model to reflect the values of liberal democ

reflect the values of liberal democ racies I'm biased I like democracy and I

racies I'm biased I like democracy and I think that if democracies inflict this

think that if democracies inflict this self-inflicted wound of shutting down

self-inflicted wound of shutting down our open source then other countries

our open source then other countries will step into this um to the to the

will step into this um to the to the detriment of of democracy so I hope that

detriment of of democracy so I hope that one of the best ways to promote

one of the best ways to promote Innovation and competition we to protect

Innovation and competition we to protect open Souls I hope all of us will do that

open Souls I hope all of us will do that thank you great

thanks well that was that was great thank you so much and you've you've

thank you so much and you've you've really laid out a landscape of a lot of

really laid out a landscape of a lot of the things that people are worried about

the things that people are worried about and talking about um so when when you

and talking about um so when when you think about maybe I'll just pick up

think about maybe I'll just pick up where you left off um you mentioned that

where you left off um you mentioned that you thought other countries might do

you thought other countries might do something I think that relates a little

something I think that relates a little bit to the economist language of

bit to the economist language of incremental risk and incremental benefit

incremental risk and incremental benefit that we were just talking about with

that we were just talking about with Percy um can you talk through a little

Percy um can you talk through a little bit uh you know counterfactual world I

bit uh you know counterfactual world I guess you know what what happens without

guess you know what what happens without the some of these open weight models and

the some of these open weight models and and can you just flush out a little bit

and can you just flush out a little bit more what you might see Happening by Bad

more what you might see Happening by Bad actors or other countries or um and what

actors or other countries or um and what might happen in in our ecosystem yeah so

might happen in in our ecosystem yeah so let's see um I think that choke points

let's see um I think that choke points will arise um I I I see two

will arise um I I I see two possibilities both bad I think one would

possibilities both bad I think one would be if everyone is beholden to a very

be if everyone is beholden to a very very small number of foundation model

very small number of foundation model providers um uh that's creating a new CH

providers um uh that's creating a new CH Point um if we don't have access to open

Point um if we don't have access to open source I think Academia will suffer a

source I think Academia will suffer a lot of the work that pery talked about

lot of the work that pery talked about and that frankly many universities all

and that frankly many universities all around the world are doing I was

around the world are doing I was actually just in career couple weeks ago

actually just in career couple weeks ago at the government meeting prime minister

at the government meeting prime minister was there and I think that also you know

was there and I think that also you know I was hearing about Korean academics

I was hearing about Korean academics saying please protect open source

saying please protect open source because academics really need to to to

because academics really need to to to monitor uh uh to to to to do their work

monitor uh uh to to to to do their work so I think Academia would have a much

so I think Academia would have a much harder time and then a lot of innovation

harder time and then a lot of innovation just won't happen because I see so many

just won't happen because I see so many startups all around the world not just

startups all around the world not just in s all around the world um building on

in s all around the world um building on top of the open models and doing things

top of the open models and doing things that are simply not possible on top of

that are simply not possible on top of the Clos models maybe one example um

the Clos models maybe one example um llama llama 3 was released as open

llama llama 3 was released as open weights um and so you know llama 3 turns

weights um and so you know llama 3 turns out has a relatively short uh uh input

out has a relatively short uh uh input context length they can only accept a

context length they can only accept a relatively short input and so some team

relatively short input and so some team you know extended that to accept a much

you know extended that to accept a much longer input expanded context length and

longer input expanded context length and if lmer 3 had been released as an API

if lmer 3 had been released as an API only thing this work would have been

only thing this work would have been impossible um and so open source Builds

impossible um and so open source Builds on top of itself as a very rich

on top of itself as a very rich ecosystem that I think some companies

ecosystem that I think some companies through really intense lobing efforts

through really intense lobing efforts are really eager they try to shut off

are really eager they try to shut off and by the way and let me say so some

and by the way and let me say so some large companies have been saying stuff

large companies have been saying stuff like oh we totally support open source

like oh we totally support open source just not of the most advanced models we

just not of the most advanced models we got to keep that away from our

got to keep that away from our adversaries it's been over the last year

adversaries it's been over the last year it's been fascinating I've not spent a

it's been fascinating I've not spent a lot of time in DC I tend to stay stay

lot of time in DC I tend to stay stay here and just write software it's been

here and just write software it's been fascinating to see how the adversaries

fascinating to see how the adversaries of Open Source have shifted their

of Open Source have shifted their arguments multiple times um it was

arguments multiple times um it was initially AI could cuse all you know AI

initially AI could cuse all you know AI is like nuclear weapons analogy that

is like nuclear weapons analogy that makes no sense AI is intelligent nuclear

makes no sense AI is intelligent nuclear weapons P sub CID the strange analogy

weapons P sub CID the strange analogy when that argument lost credibility it

when that argument lost credibility it was then oh AI could create bioweapons

was then oh AI could create bioweapons but then um open and Rand published

but then um open and Rand published reports showing that the incremental

reports showing that the incremental risk um is very small uh so yes of

risk um is very small uh so yes of course AI could be used to create B

course AI could be used to create B weapons I'll just say that so can an

weapons I'll just say that so can an Excel spreadsheet um I mean Excel

Excel spreadsheet um I mean Excel spreadsheets I you can track experiments

spreadsheets I you can track experiments I mean Excel is really dangerous let's

I mean Excel is really dangerous let's regulate that so but AI so the evidence

regulate that so but AI so the evidence that AI helps more there is no evidence

that AI helps more there is no evidence as far as I'm aware that AI helps more

as far as I'm aware that AI helps more than you know an Excel spreadsheet or

than you know an Excel spreadsheet or web search for for creating bioweapons

web search for for creating bioweapons and then people have studied this so

and then people have studied this so when that credibility then it became oh

when that credibility then it became oh you know our adversaries China

you know our adversaries China specifically like got to got to keep

specifically like got to got to keep maybe Russia I don't know right so it's

maybe Russia I don't know right so it's been fascinating watching the shifting

been fascinating watching the shifting arguments of the of the open open source

arguments of the of the open open source and I've also been fascinated it's been

and I've also been fascinated it's been amazing to watch how effective some of

amazing to watch how effective some of these companies have been at convincing

these companies have been at convincing governments to regulate very clearly

governments to regulate very clearly against uh uh their own national

against uh uh their own national interests this this more mainly outside

interests this this more mainly outside the United States

the United States but thank you so much um well and so you

but thank you so much um well and so you know one

know one we're so fortunate to have you here from

we're so fortunate to have you here from so many different perspectives you have

so many different perspectives you have because of course you did foundational

because of course you did foundational research as a computer science professor

research as a computer science professor at Stanford um you cound co-founded

at Stanford um you cound co-founded corsera uh so anybody wants to learn

corsera uh so anybody wants to learn about this um you can watch his videos

about this um you can watch his videos um and you're now an also an investor

um and you're now an also an investor and an entrepreneur um can I turn this

and an entrepreneur um can I turn this around a little bit to think about the

around a little bit to think about the investor perspective if you are

investor perspective if you are investing in applications um how do you

investing in applications um how do you think about possible choke points and

think about possible choke points and risks that your Investments might um

risks that your Investments might um might experience yeah I think the the

might experience yeah I think the the the choke point that we all love to

the choke point that we all love to avoid a lot of it is at the um

avoid a lot of it is at the um Foundation model layer uh uh and I think

Foundation model layer uh uh and I think they'll be hard for um companies to

they'll be hard for um companies to establish a CH point there if we promote

establish a CH point there if we promote and invest in open source there's a

and invest in open source there's a government that does having

government that does having conversations with the where the

conversations with the where the administration was considering for their

administration was considering for their own procurement process to really have a

own procurement process to really have a bias fail colting open source based

bias fail colting open source based Solutions I don't know if they'll do it

Solutions I don't know if they'll do it but you know one government not the US

but you know one government not the US that that was s do was was considering

that that was s do was was considering this I think that type of move to

this I think that type of move to promote open source would very good um

promote open source would very good um there's one other layer that may to

there's one other layer that may to comment on that's kind of emerging which

comment on that's kind of emerging which is on top of the foundation model layer

is on top of the foundation model layer there's also another layer that's

there's also another layer that's emerging called an orchestration layer

emerging called an orchestration layer and what that means is um uh often um

and what that means is um uh often um when you're Building Systems you don't

when you're Building Systems you don't just call an L large language model once

just call an L large language model once you call it a bunch of times and you

you call it a bunch of times and you call it once as output goes into the

call it once as output goes into the second prom goes to the third prom you

second prom goes to the third prom you pull some data toloy a rag and so on so

pull some data toloy a rag and so on so that orchestration layer um right now is

that orchestration layer um right now is very Dynamic very competitive so there

very Dynamic very competitive so there is no CH point there at this moment but

is no CH point there at this moment but you know I definitely see uh more

you know I definitely see uh more startups at this point prior trying to

startups at this point prior trying to trying to become a dominant player but

trying to become a dominant player but that's very good right you you want you

that's very good right you you want you know um uh entrepreneurs with low

know um uh entrepreneurs with low resources aspiring to build the next

resources aspiring to build the next Google or the next you know mea or

Google or the next you know mea or something so I think it's very intense

something so I think it's very intense competition uh but I I could see a

competition uh but I I could see a future many many years from now if if

future many many years from now if if there's a new player that that that wins

there's a new player that that that wins so could you I'd love to hear you

so could you I'd love to hear you elaborate a little bit about that

elaborate a little bit about that because I know that you've talked about

because I know that you've talked about like AI agents as the next big wave in

like AI agents as the next big wave in applications and I think that kind of

applications and I think that kind of picks up on this orchestration layer and

picks up on this orchestration layer and I think you know most people in the

I think you know most people in the audience have seen the images they've

audience have seen the images they've seen the text they've played with the

seen the text they've played with the text but um it's these it's it's a

text but um it's these it's it's a little bit harder for the ordinary

little bit harder for the ordinary person to Intuit it like what's going on

person to Intuit it like what's going on behind the scenes and also frankly

behind the scenes and also frankly what's possible when you have multiple

what's possible when you have multiple agents kind of interacting together so

agents kind of interacting together so I'm wondering if you could talk us

I'm wondering if you could talk us through what makes you excited about

through what makes you excited about that or maybe a use case or an example

that or maybe a use case or an example if you have one sure yeah so I think one

if you have one sure yeah so I think one of the most exciting Trends in Tech

of the most exciting Trends in Tech right now is AI agents or agentic Ai

right now is AI agents or agentic Ai workflows and and lcy here's what I mean

workflows and and lcy here's what I mean um many of us are used to going into a

um many of us are used to going into a large langage model and prompting it uh

large langage model and prompting it uh to have it write a response right and

to have it write a response right and that's kind of like imagine you go to

that's kind of like imagine you go to someone and say I like you to write a

someone and say I like you to write a topic i' like you to write an essay on a

topic i' like you to write an essay on a certain topic but I need you to type

certain topic but I need you to type from start to finish without ever

from start to finish without ever hitting backspace or without ever going

hitting backspace or without ever going back to edit your work um or if you ask

back to edit your work um or if you ask AI to write code is like going to

AI to write code is like going to developer and saying I need you to just

developer and saying I need you to just type out a computer program and just

type out a computer program and just type it out from start to finish and

type it out from start to finish and then I want it to work and it turns out

then I want it to work and it turns out that you know despite the difficulty of

that you know despite the difficulty of this right like I can't write like that

this right like I can't write like that I can't cope like that AI actually does

I can't cope like that AI actually does surprisingly well at this task

surprisingly well at this task but it turns out that if you use an

but it turns out that if you use an agentic AI workflow in which you can ask

agentic AI workflow in which you can ask the AI write an outline and then think

the AI write an outline and then think about your outline do you need to do any

about your outline do you need to do any web research if so what do you want to

web research if so what do you want to type into the search bar a web search

type into the search bar a web search engine then go do a bunch of you know

engine then go do a bunch of you know Google or Bing or duck go searches or

Google or Bing or duck go searches or whatever and then get the results and

whatever and then get the results and then write the first draft and then read

then write the first draft and then read your own draft and think about where you

your own draft and think about where you want to improve it and iterate over your

want to improve it and iterate over your own work over and over so that's called

own work over and over so that's called an agent take workflow and it turns out

an agent take workflow and it turns out that this does way better than if you

that this does way better than if you force an AI system to write from start

force an AI system to write from start to finish um one the exciting things I

to finish um one the exciting things I found is that uh my team collected data

found is that uh my team collected data the kind of third party data uh sorry we

the kind of third party data uh sorry we we did a survey of uh uh AI uh large

we did a survey of uh uh AI uh large language models and agents applied to

language models and agents applied to writing code this a standard Benchmark

writing code this a standard Benchmark from open AI called human eval that

from open AI called human eval that measures how well AI systems right code

measures how well AI systems right code and it turns out that chat GPT or GPT

and it turns out that chat GPT or GPT 3.5 um had 48% right in term on on this

3.5 um had 48% right in term on on this Benchmark GPT 4 had I think I want to

Benchmark GPT 4 had I think I want to say 67% 67 or 68% um so gbd4 was a huge

say 67% 67 or 68% um so gbd4 was a huge improvement from 48 to 67 68% but it

improvement from 48 to 67 68% but it turns out the improvement from 3.5 gbd

turns out the improvement from 3.5 gbd 3.5 to gbd4 is dwarfed by the

3.5 to gbd4 is dwarfed by the improvement from if you wrap an agentic

improvement from if you wrap an agentic workflow around this and even using qbd

workflow around this and even using qbd 3.5 on this Benchmark can get kind of up

3.5 on this Benchmark can get kind of up to 95% accuracy um uh by having the AI

to 95% accuracy um uh by having the AI system write code and it run some tests

system write code and it run some tests fix his own code and just iterate over

fix his own code and just iterate over and over to keep on improving his code

and over to keep on improving his code until it works so I think for those of

until it works so I think for those of you interested about applications I

you interested about applications I think agentic AI um is going to be a

think agentic AI um is going to be a very large Trend that that I think we're

very large Trend that that I think we're actually seeing this in the field

actually seeing this in the field already my team AI fund we we we built

already my team AI fund we we we built startups we built one startup per month

startups we built one startup per month but for many applications um you know in

but for many applications um you know in legal work and Healthcare and tons of

legal work and Healthcare and tons of other applications we seeing these

other applications we seeing these agentic workflows um deliver really

agentic workflows um deliver really fantastic results that we couldn't with

fantastic results that we couldn't with other ways and so what else could it do

other ways and so what else could it do besides like search the web which again

besides like search the web which again I think we we we have some intuition for

I think we we we have some intuition for that but if you were building a custom

that but if you were building a custom one of those what are other kinds of

one of those what are other kinds of calls that that this type of agentic AI

calls that that this type of agentic AI could could make yeah so maybe some some

could could make yeah so maybe some some techniques I find exciting the gener ref

techniques I find exciting the gener ref one is reflection where we ask you to

one is reflection where we ask you to read over your own output and just see

read over your own output and just see if you have any ways to improve it that

if you have any ways to improve it that surprisingly helps quite a bit um to use

surprisingly helps quite a bit um to use where you can search the web or it can

where you can search the web or it can um search different information sources

um search different information sources or access your calendar with permission

or access your calendar with permission if that's what you needed to do or

if that's what you needed to do or sometimes it can also um make an API

sometimes it can also um make an API call to take an action so you know

call to take an action so you know people are building agents uh customer

people are building agents uh customer service agents for example they can make

service agents for example they can make an API call to retrieve information or

an API call to retrieve information or even make an API call to maybe issue a

even make an API call to maybe issue a refund you know um uh and then I think

refund you know um uh and then I think uh M uh and then agents can increasingly

uh M uh and then agents can increasingly plan out complex sequence of of steps

plan out complex sequence of of steps where we can say to do this time I need

where we can say to do this time I need to First do research then write an

to First do research then write an outline then fact check and then write

outline then fact check and then write the final jop or whatever it can decide

the final jop or whatever it can decide for itself what sequence of actions to

for itself what sequence of actions to take and then lost the multi-agent

take and then lost the multi-agent collaborations has been a fascinating

collaborations has been a fascinating Trend where it turns out that if you

Trend where it turns out that if you have two AI systems and you say you are

have two AI systems and you say you are the writer you're the researcher you're

the writer you're the researcher you're the fact Checker and then build three

the fact Checker and then build three agents uh and each agent is really built

agents uh and each agent is really built by prompting an OM you tell the dear

by prompting an OM you tell the dear your role is you are now a fact Checker

your role is you are now a fact Checker so go fact check these please well dear

so go fact check these please well dear you're now a writer so go write a clear

you're now a writer so go write a clear article uh then if you have multi-agents

article uh then if you have multi-agents interact that also uh really tends to

interact that also uh really tends to drive you know improved results so and

drive you know improved results so and it's a very nice and very exciting space

it's a very nice and very exciting space right frankly six years ago this stuff

right frankly six years ago this stuff it like kind of wasn't working six a

it like kind of wasn't working six a year ago it was demo whereare it was

year ago it was demo whereare it was great demos really couldn't get to work

great demos really couldn't get to work at all but now I feel like I'm seeing

at all but now I feel like I'm seeing month over month improvements on agentic

month over month improvements on agentic AI workflows um so this is actually a

AI workflows um so this is actually a very exciting space and and I think um

very exciting space and and I think um uh it does to hyper competitive right

uh it does to hyper competitive right now at the agentic layer and also the uh

now at the agentic layer and also the uh Innovation all the applications how do

Innovation all the applications how do you take this and apply this to you know

you take this and apply this to you know many different domains that's a very ex

many different domains that's a very ex that that layer of innovation feels like

that that layer of innovation feels like is exploding as well very exciting time

is exploding as well very exciting time to be in so one of the things that was

to be in so one of the things that was interesting about the the first large

interesting about the the first large language models is that they were mostly

language models is that they were mostly using open data that they scraped and

using open data that they scraped and later this afternoon we'll we're going

later this afternoon we'll we're going to come back and talk more about the

to come back and talk more about the scraping issue but they they were using

scraping issue but they they were using kind of common data and not as much like

kind of common data and not as much like proprietary data not as much like user

proprietary data not as much like user cck data or things like that but um you

cck data or things like that but um you know in some of these applications

know in some of these applications you're talking about and also I guess

you're talking about and also I guess previously the foundation models would

previously the foundation models would would kind of you know interact with

would kind of you know interact with lots of different things but they

lots of different things but they weren't they didn't they weren't like

weren't they didn't they weren't like directly connecting to other software

directly connecting to other software products so the agentic products you're

products so the agentic products you're talking about you mentioned they might

talking about you mentioned they might read your email um we later this

read your email um we later this afternoon we we'll we're going to talk

afternoon we we'll we're going to talk about um you know using like crms

about um you know using like crms talking to crms so some of these layers

talking to crms so some of these layers might now need to interoperate with

might now need to interoperate with other software products and they might

other software products and they might also possibly you know make more use of

also possibly you know make more use of user feedback data depending on what

user feedback data depending on what they're interoperating with so I'm sort

they're interoperating with so I'm sort of curious if you could do a little bit

of curious if you could do a little bit of like forward looking where you things

of like forward looking where you things are think things are going um what are

are think things are going um what are what are some of the inputs and

what are some of the inputs and bottlenecks that might come up in the

bottlenecks that might come up in the future implementations of these as they

future implementations of these as they start interoperating with more systems

start interoperating with more systems you know it's really interesting I I I

you know it's really interesting I I I really appreciate I think Susan has

really appreciate I think Susan has always been a very long-term forward

always been a very long-term forward thinker I have to say I find myself

thinker I have to say I find myself inspired by that view and honestly s not

inspired by that view and honestly s not I'll get to your question honestly what

I'll get to your question honestly what I'm really worried about is Les the

I'm really worried about is Les the longterm future is the awful regulations

longterm future is the awful regulations I'm seeing right now um that that just

I'm seeing right now um that that just as an example California s SP 1047 I

as an example California s SP 1047 I think it' be awful for American

think it' be awful for American innovation and for California inov I

innovation and for California inov I can't believe that California home of

can't believe that California home of such an Innovative place with so many

such an Innovative place with so many inventions is contemplating such

inventions is contemplating such ridiculous regulation is equivalent of

ridiculous regulation is equivalent of saying if your electric motor is used by

saying if your electric motor is used by anyone for an inferious purpose you

anyone for an inferious purpose you could be liable you know it's just awful

could be liable you know it's just awful and I don't see how the state of

and I don't see how the state of California could could seriously

California could could seriously contemplate something like that um sorry

contemplate something like that um sorry but to take a longer longer looking view

but to take a longer longer looking view you know I I I feel like um uh uh I

you know I I I feel like um uh uh I think that one interesting Dynamic would

think that one interesting Dynamic would be where will incumbents win and where

be where will incumbents win and where will new entrance have a shot so I think

will new entrance have a shot so I think um in the case of electronic health

um in the case of electronic health records you know I think we know that

records you know I think we know that epic as electronic health record system

epic as electronic health record system right has massive Market power immense

right has massive Market power immense concentration um so where will

concentration um so where will incumbents like that be able to lock

incumbents like that be able to lock down options to to to block innovators

down options to to to block innovators um I think I think those are those are

um I think I think those are those are issues worth thinking about

issues worth thinking about um yeah I don't know I I I think sorry

um yeah I don't know I I I think sorry yeah I I maybe part of me feels like for

yeah I I maybe part of me feels like for a lot of applications it feels like a

a lot of applications it feels like a relatively Dynamic landscape still um

relatively Dynamic landscape still um but I think access to data is going to

but I think access to data is going to be a thing um and I think I I I worry a

be a thing um and I think I I I worry a little bit about the um some of the

little bit about the um some of the emerging propriety deuse to access data

emerging propriety deuse to access data um one one of the debates is what should

um one one of the debates is what should fair use be in the era of of AI is it

fair use be in the era of of AI is it okay for a company to read something on

okay for a company to read something on the open internet and use it to train

the open internet and use it to train the AI model U my personal preference I

the AI model U my personal preference I believe the answer should be yes but

believe the answer should be yes but that's my kind of personal opinion um

that's my kind of personal opinion um and I realized that so I post on social

and I realized that so I post on social media so you uh New York Times suit open

media so you uh New York Times suit open AI uh and Microsoft I think and I found

AI uh and Microsoft I think and I found myself more on open AI side than on than

myself more on open AI side than on than on the New York Times side in that

on the New York Times side in that matter because um and and I think I

matter because um and and I think I learned an interesting lesson when I

learned an interesting lesson when I post on social media second said you

post on social media second said you know I as a human am allowed to read

know I as a human am allowed to read articles on the open internet and learn

articles on the open internet and learn from it I would like AI to be able to do

from it I would like AI to be able to do so too and I got a bit of you know hate

so too and I got a bit of you know hate SL criticism um including from some uh

SL criticism um including from some uh Sanford faculty um uh saying that I had

Sanford faculty um uh saying that I had no reason for making an equivalence

no reason for making an equivalence between humans and AI right because the

between humans and AI right because the point was yes humans can read stuff on

point was yes humans can read stuff on the internet but just because a human

the internet but just because a human can do it why is there the reason to let

can do it why is there the reason to let AI do it and I realized that

AI do it and I realized that I think there two views to AI um I tend

I think there two views to AI um I tend to view AI as an extension of myself I I

to view AI as an extension of myself I I think of as a tool like I use a web

think of as a tool like I use a web browser to do this and someone may use

browser to do this and someone may use accessibility software to make the font

accessibility software to make the font bigger or you know or just display a web

bigger or you know or just display a web page and high contract so to me that's a

page and high contract so to me that's a tool I'm going to use AI to do for me

tool I'm going to use AI to do for me something that I could do and so if I

something that I could do and so if I could do something why can't I use a

could do something why can't I use a tool to do it too I can read web pages I

tool to do it too I can read web pages I want to use a tool to do it more

want to use a tool to do it more efficiently and I realize there are

efficiently and I realize there are others I'm blessed with that view that

others I'm blessed with that view that view AI is almost a different species

view AI is almost a different species this this thing separate from themselves

this this thing separate from themselves and I don't think animals need to

and I don't think animals need to necessarily have the same rights as

necessarily have the same rights as humans you know I think human I mean I

humans you know I think human I mean I value human life and I mean you know

value human life and I mean you know sometimes if a bug is annoying me you

sometimes if a bug is annoying me you know I will not treat it very nicely so

know I will not treat it very nicely so I don't think I don't think all bugs

I don't think I don't think all bugs need to have the same I don't think

need to have the same I don't think other species you know need to have

other species you know need to have exact same rights as as humans and I

exact same rights as as humans and I think realiz there's some people that

think realiz there's some people that view AI as almost a separate species and

view AI as almost a separate species and so we don't need to give them the same

so we don't need to give them the same right to read the internet as as humans

right to read the internet as as humans do so I I tend to view AI as a tool and

do so I I tend to view AI as a tool and that's why I took a certain view but was

that's why I took a certain view but was interesting seeing you know these two

interesting seeing you know these two different uh so this is a very

different uh so this is a very interesting conversation and we I just

interesting conversation and we I just want to advertise uh that we will have a

want to advertise uh that we will have a panel at the uh later this afternoon and

panel at the uh later this afternoon and we will hear from the authors and

we will hear from the authors and Publishers um who probably have a a

Publishers um who probably have a a different view from you and uh and that

different view from you and uh and that will be that will be fleshed out um this

will be that will be fleshed out um this this conversation is one that is clearly

this conversation is one that is clearly extremely important to be had and you

extremely important to be had and you know the perspective of that is an input

know the perspective of that is an input to this open source which is you know

to this open source which is you know providing all of this Innovation seems

providing all of this Innovation seems like you know that's that's a a a

like you know that's that's a a a perspective um to be weighed yeah and

perspective um to be weighed yeah and just to say I I do acknowledge this is a

just to say I I do acknowledge this is a difficult issue I honestly don't know

difficult issue I honestly don't know what's the my gut was let's let AI read

what's the my gut was let's let AI read the open internet but I do think it's a

the open internet but I do think it's a difficult issue I don't think I I don't

difficult issue I don't think I I don't claim to know the right answer and I

claim to know the right answer and I think protecting the Free Press which is

think protecting the Free Press which is a key pillar of democracy I think there

a key pillar of democracy I think there are very difficult issues I I don't

are very difficult issues I I don't think I know the full answer to so very

think I know the full answer to so very good well one of the few things you may

good well one of the few things you may not know the full answer too uh we want

not know the full answer too uh we want we always wonder about the AI in your

we always wonder about the AI in your brain um well just in the last couple

brain um well just in the last couple minutes um our next panel after the

minutes um our next panel after the break is going to be about um Healthcare

break is going to be about um Healthcare and Ai and I know you also have a long

and Ai and I know you also have a long history with health applications you

history with health applications you mentioned um access to electronic

mentioned um access to electronic medical record data um which could be a

medical record data um which could be a a bottleneck here but I'm curious

a bottleneck here but I'm curious generally um you know what are other

generally um you know what are other ways potentially completely different

ways potentially completely different ways that you might see AI impacting the

ways that you might see AI impacting the health care industry and sort of what

health care industry and sort of what are the other inputs that are needed for

are the other inputs that are needed for this health Innovation you know it's

this health Innovation you know it's been so so um my my my group here at St

been so so um my my my group here at St has done a lot of work on Healthcare we

has done a lot of work on Healthcare we deployed stuff at St Hospital you kind

deployed stuff at St Hospital you kind of publishing a bunch of papers and so

of publishing a bunch of papers and so on it's been fascinating watching how

on it's been fascinating watching how difficult it is to approach Healthcare

difficult it is to approach Healthcare in the US um I feel like um and in my

in the US um I feel like um and in my team of AI fun we see opportunities left

team of AI fun we see opportunities left and right right there are so many ways

and right right there are so many ways we can drive efficiencies improve

we can drive efficiencies improve patient care make doctor's lives better

patient care make doctor's lives better and the problem we keep on running it

and the problem we keep on running it over and over is not that we don't know

over and over is not that we don't know how to use AI to make things better it's

how to use AI to make things better it's the Golden Market you know I think that

the Golden Market you know I think that with the US patient payer provider

with the US patient payer provider missign incentives lots of Choke points

missign incentives lots of Choke points how do you how do you get a how do you

how do you how do you get a how do you get a payer to you know how do you how

get a payer to you know how do you how do you get that to be a code so the

do you get that to be a code so the payer can reimburse and all that so

payer can reimburse and all that so those tend to be the problems not even

those tend to be the problems not even the technical problems or how the build

the technical problems or how the build things that can actually make lives

things that can actually make lives better um and then the the other Dynamic

better um and then the the other Dynamic I'm seeing is I'm seeing um uh because

I'm seeing is I'm seeing um uh because of this narly ball of stuff that's so

of this narly ball of stuff that's so hard to approach in the US I'm seeing a

hard to approach in the US I'm seeing a lot of um Innovation and and and

lot of um Innovation and and and startups Pro profit nonprofit academic

startups Pro profit nonprofit academic happening overseas rather than in the

happening overseas rather than in the United States uh was visiting Singapore

United States uh was visiting Singapore a few months ago I was seeing stuff

a few months ago I was seeing stuff deployed in hospitals in Singapore that

deployed in hospitals in Singapore that I've not seen deploy anywhere in the

I've not seen deploy anywhere in the United States so you know check on the

United States so you know check on the hospital in Singapore if you have pum or

hospital in Singapore if you have pum or read EHR and try to give the doctor

read EHR and try to give the doctor guidance on you know this do does it see

guidance on you know this do does it see something that maybe the doctor has

something that maybe the doctor has missed I see stuff deployed in Singapore

missed I see stuff deployed in Singapore which has a unified patient ID basically

which has a unified patient ID basically uh that we don't have in the US you know

uh that we don't have in the US you know low litigation

low litigation risks all sorts of things and then

risks all sorts of things and then recently aian recently working with a

recently aian recently working with a with a startup that decided to um pursue

with a startup that decided to um pursue India Golden Market because candidly

India Golden Market because candidly it's going to be much easier to take

it's going to be much easier to take certain things to Market in India than

certain things to Market in India than here in the United States and just as a

here in the United States and just as a as a you know as as a as a US citizen I

as a you know as as a as a US citizen I I just felt kind of sad that we can't

I just felt kind of sad that we can't seem to get the job done here in America

seem to get the job done here in America um well so thank you so much Andrew you

um well so thank you so much Andrew you your many different perspectives in here

your many different perspectives in here were so interesting and I also love that

were so interesting and I also love that you're not afraid to say what you think

you're not afraid to say what you think um so that it makes for a much more

um so that it makes for a much more powerful and interesting conversation

powerful and interesting conversation can I can I add one comment that might

can I can I add one comment that might gave into trouble okay all right so I I

gave into trouble okay all right so I I know that a lot government officials

know that a lot government officials here so this is I should say but

here so this is I should say but honestly I'm I'm I'm I'm Pro World prot

honestly I'm I'm I'm I'm Pro World prot Tech and I'm also pro-america one thing

Tech and I'm also pro-america one thing that saddened me was it felt like um uh

that saddened me was it felt like um uh it felt like pava governments have

it felt like pava governments have become anti-tech in a way that I've not

become anti-tech in a way that I've not seen before so in previous ad multiple

seen before so in previous ad multiple previous administrations not one

previous administrations not one multiple previous administrations if

multiple previous administrations if European Regulators were messing around

European Regulators were messing around with American companies our government

with American companies our government went in to argue the Europeans right and

went in to argue the Europeans right and I appreciated that now I kind of see

I appreciated that now I kind of see when you know foreign Regulators are

when you know foreign Regulators are messing with American companies feels

messing with American companies feels like sometimes our Administration

like sometimes our Administration doesn't lift a finger to help them um

doesn't lift a finger to help them um I'm sure I'm seeing only isolated cases

I'm sure I'm seeing only isolated cases I'm sure this is a broader you know

I'm sure this is a broader you know context I'm not seeing but I feel like

context I'm not seeing but I feel like I'm prot Tech I'm I'm Pro competition

I'm prot Tech I'm I'm Pro competition I'm Pro startup I'm also pro-american

I'm Pro startup I'm also pro-american big Tech um and and I wish that uh uh

big Tech um and and I wish that uh uh SLI Valley and America American Tech in

SLI Valley and America American Tech in general is very precious and then candid

general is very precious and then candid when I see politicians call up

when I see politicians call up individual

individual entrepreneurs Like Bernie Sanders call

entrepreneurs Like Bernie Sanders call up Elon Musk by name for the sin of

up Elon Musk by name for the sin of being successful and Wealthy you know

being successful and Wealthy you know and yes he's committed other sins but it

and yes he's committed other sins but it feels like it feels like a I wish I wish

feels like it feels like a I wish I wish I felt like America remained you know

I felt like America remained you know firmly Pro competition and Pro Tech

firmly Pro competition and Pro Tech including big Tech and small Tech and St

including big Tech and small Tech and St but let's let's just keep building this

but let's let's just keep building this together um yeah what so Andrew thank

together um yeah what so Andrew thank you so much for being here we're going

you so much for being here we're going to move to a coffee break and we'll come

to move to a coffee break and we'll come back in about 10 minutes for the next

back in about 10 minutes for the next session thank

session thank [Applause]

good morning everyone we're going to get started again uh so if I could ask

started again uh so if I could ask everyone to take their

seats I'm Jennifer Dixon I'm an assistant chief in our policy section at

assistant chief in our policy section at the antitrust Division and also special

the antitrust Division and also special counsel for IP and policy and I will be

counsel for IP and policy and I will be one of your MC's of the day so if you

one of your MC's of the day so if you have any questions please do uh reach

have any questions please do uh reach out to me um now that we've heard from

out to me um now that we've heard from uh Professor uh leang and Professor ing

uh Professor uh leang and Professor ing uh about Foundation models they set the

uh about Foundation models they set the stage about open and closed uh models

stage about open and closed uh models the benefits of of uh open- Source

the benefits of of uh open- Source models we've learned about applications

models we've learned about applications we're going to now apply that to our

we're going to now apply that to our discussion of healthc care applications

discussion of healthc care applications and I wanted to introduce our uh

and I wanted to introduce our uh moderator uh Katrina Rouse who is a a

moderator uh Katrina Rouse who is a a colleague of mine at the antitrust

colleague of mine at the antitrust division uh she is the director of the

division uh she is the director of the antitrust division's new task force on

antitrust division's new task force on healthc care uh she also is the deputy

healthc care uh she also is the deputy director of civil enforcement um and a

director of civil enforcement um and a special counsel for healthcare so thank

special counsel for healthcare so thank you Katrina and she will introduce the

you Katrina and she will introduce the rest of our panelists and all of their

rest of our panelists and all of their full bios are um in the program

full bios are um in the program materials so go ahead

materials so go ahead thanks very much Jennifer I am thrilled

thanks very much Jennifer I am thrilled to moderate this Healthcare Spotlight

to moderate this Healthcare Spotlight panel where we'll discuss the current

panel where we'll discuss the current state of competition in the healthare AI

state of competition in the healthare AI space the role that data access and

space the role that data access and Advantage plays in this competition what

Advantage plays in this competition what the dimensions of competition in this

the dimensions of competition in this space are and could be and the potential

space are and could be and the potential for algorithmic collusion among other

for algorithmic collusion among other topics I am really looking forward to a

topics I am really looking forward to a fascinating conversation today with

fascinating conversation today with these five exceptional panelists and

these five exceptional panelists and I'll now introduce them Elena Vach is a

I'll now introduce them Elena Vach is a partner at the VC fund General Catalyst

partner at the VC fund General Catalyst where she focuses on investing in life

where she focuses on investing in life sciences she serves on several boards

sciences she serves on several boards and prior to joining General Catalyst

and prior to joining General Catalyst she helped build the soft Bank Vision

she helped build the soft Bank Vision funds Healthcare

funds Healthcare portfolio David kizner is the general

portfolio David kizner is the general counsel and Chief privacy officer at viz

counsel and Chief privacy officer at viz AI a leading AI care coordination

AI a leading AI care coordination platform for disease detection and

platform for disease detection and workflow optimization prior to viai

workflow optimization prior to viai David served in legal roles for Stanford

David served in legal roles for Stanford University Stanford Healthcare Phillips

University Stanford Healthcare Phillips Healthcare and

Healthcare and Center next we have Z overmire he's the

Center next we have Z overmire he's the Blue Cross of California distinguished

Blue Cross of California distinguished associate professor of Health policy and

associate professor of Health policy and management at the UC Berkeley School of

management at the UC Berkeley School of Public Health where he does research at

Public Health where he does research at the intersection of machine learning

the intersection of machine learning medicine and health policy he is a

medicine and health policy he is a practicing emergency medicine physician

practicing emergency medicine physician and he's also a Founder he co-founded

and he's also a Founder he co-founded dandelion health an AI Innovation

dandelion health an AI Innovation platform that makes Healthcare data

platform that makes Healthcare data available to algorithm developers and he

available to algorithm developers and he founded a nonprofit Nightingale open

founded a nonprofit Nightingale open science which builds out data sets in

science which builds out data sets in partnership with health

partnership with health systems then we have Ben handle he's an

systems then we have Ben handle he's an associate professor of Economics at UC

associate professor of Economics at UC Berkeley and a faculty research fellow

Berkeley and a faculty research fellow at the National Bureau of economic

at the National Bureau of economic research his important research focuses

research his important research focuses on the microeconomics of consumer choice

on the microeconomics of consumer choice and Market structure in the healthcare

and Market structure in the healthcare sector with an emphasis on health

sector with an emphasis on health insurance

insurance markets and then finally We Have Allison

markets and then finally We Have Allison oliger she is the chief data officer and

oliger she is the chief data officer and director of The Office of Enterprise

director of The Office of Enterprise data and analytics at the centers for

data and analytics at the centers for Medicare and Medicaid services is in

Medicare and Medicaid services is in this role Allison oversees cms's data

this role Allison oversees cms's data and information product portfolio and

and information product portfolio and she manages the development advance of

she manages the development advance of advanced analytics using CMS

advanced analytics using CMS data so to begin I thought we could do a

data so to begin I thought we could do a quick overview of the major categories

quick overview of the major categories of healthcare AI applications I think we

of healthcare AI applications I think we all understand that where there is huge

all understand that where there is huge amounts of Health Data um we should be

amounts of Health Data um we should be able to use algorithms to make better

able to use algorithms to make better medical decisions but I'd really like to

medical decisions but I'd really like to drill down on the specifics here and I

drill down on the specifics here and I thought David maybe you could start by

thought David maybe you could start by giving us an overview and examples of

giving us an overview and examples of the clinical applications for healthcare

the clinical applications for healthcare AI including what viz AI does sure

AI including what viz AI does sure there's there's hundreds I will probably

there's there's hundreds I will probably focus on what viz does and unfortunately

focus on what viz does and unfortunately this isn't the exciting uh generative AI

this isn't the exciting uh generative AI uh this is kind of old school computer

uh this is kind of old school computer vision so um interrogating like a data

vision so um interrogating like a data set so uh CT uh studies or MRIs or

set so uh CT uh studies or MRIs or ultrasound and uh finding suspected

ultrasound and uh finding suspected conditions so the first one I'll talk

conditions so the first one I'll talk about which is um very relevant because

about which is um very relevant because we're in American stroke awareness right

we're in American stroke awareness right now is the first algorithm that viz

now is the first algorithm that viz developed called V viz lvo which stands

developed called V viz lvo which stands for large vessel occlusion so it's a

for large vessel occlusion so it's a type of stroke uh won't bore you with

type of stroke uh won't bore you with all the details but effectively there's

all the details but effectively there's two types of stroke there's a es schic

two types of stroke there's a es schic stroke which is a blockage and there was

stroke which is a blockage and there was a hemorrhagic stroke which is a bleed

a hemorrhagic stroke which is a bleed they're treated very very differently

they're treated very very differently and if you treat one like the other you

and if you treat one like the other you could kill a patient so uh there's a

could kill a patient so uh there's a saying in stroke time is brain the clock

saying in stroke time is brain the clock is ticking your brain is being starved

is ticking your brain is being starved of oxygen and for an es schic stroke a

of oxygen and for an es schic stroke a large vessal occlusion uh the doctors

large vessal occlusion uh the doctors joke they refer them to themselves as

joke they refer them to themselves as plumbers because you're literally

plumbers because you're literally clearing a block right and then you're

clearing a block right and then you're restoring blood flow and oxygen to the

restoring blood flow and oxygen to the brain the faster you do it the better

brain the faster you do it the better the outcomes you have so what does viz

the outcomes you have so what does viz lvo do uh when a patient presents

lvo do uh when a patient presents suspected of having a stroke in an

suspected of having a stroke in an emergency room they're going to get a CT

emergency room they're going to get a CT scan of their head

scan of their head you know typical workflow that would go

you know typical workflow that would go off to a radiologist to read uh they

off to a radiologist to read uh they they might you know pick it up quickly

they might you know pick it up quickly they' alert the emergency room um if

they' alert the emergency room um if that hospital is capable of doing a

that hospital is capable of doing a mechanical thrombectomy they could do it

mechanical thrombectomy they could do it if it's not they have to transfer the

if it's not they have to transfer the patient so viz um algorithm detects that

patient so viz um algorithm detects that suspected large vessal occlusion on a CT

suspected large vessal occlusion on a CT scan has a whole application for secure

scan has a whole application for secure communication and image viewing so that

communication and image viewing so that the care team can coordinate the care of

the care team can coordinate the care of that patient either get the patient

that patient either get the patient treated at that hospital or transferred

treated at that hospital or transferred as quickly as possible so that's that's

as quickly as possible so that's that's viz lvo and and having real world um

viz lvo and and having real world um Improvement in patient outcomes you know

Improvement in patient outcomes you know reducing costs for uh having better

reducing costs for uh having better Financial outcomes for Health Care

Financial outcomes for Health Care Systems and payers uh another algorithm

Systems and payers uh another algorithm I'll talk about is called viz HCM so

I'll talk about is called viz HCM so hypertrophic cardiomyopathy is a cardiac

hypertrophic cardiomyopathy is a cardiac condition um you may hear of this from

condition um you may hear of this from time to time about like an athlete in

time to time about like an athlete in the prime of their life who suddenly has

the prime of their life who suddenly has a heart attack and May die on the

a heart attack and May die on the practice field this is uh I believe it's

practice field this is uh I believe it's the number one cardiac killer of people

the number one cardiac killer of people under 30 and often times people are

under 30 and often times people are asymptomatic they don't even know that

asymptomatic they don't even know that they have this disease so viz um in

they have this disease so viz um in partnership with Bristol Meer squib

partnership with Bristol Meer squib developed an algorithm that detects

developed an algorithm that detects suspected HCM and just a simple 12 lead

suspected HCM and just a simple 12 lead ECG and again it's not acute like stroke

ECG and again it's not acute like stroke where the the time is really ticking but

where the the time is really ticking but it gets those um alerts to the right

it gets those um alerts to the right clinicians who can take action and help

clinicians who can take action and help the patient

the patient um and then I'll mention one that is uh

um and then I'll mention one that is uh generative Ai and probably the folks

generative Ai and probably the folks from Stanford who who work on on this

from Stanford who who work on on this know more about it but um when I was

know more about it but um when I was here at Stanford uh uh Stanford um is

here at Stanford uh uh Stanford um is beginning to use AI to help clinicians

beginning to use AI to help clinicians do a first draft of a response when

do a first draft of a response when patients interact with them over the My

patients interact with them over the My Health application which is part of Epic

Health application which is part of Epic um so the clinicians are intended to

um so the clinicians are intended to then you know review that but it helps

then you know review that but it helps them um respond to the overwhelming

them um respond to the overwhelming number of inquiries they get from

number of inquiries they get from patients like us so those are those are

patients like us so those are those are three examples in the healthcare space

three examples in the healthcare space can I hop in with a little bit about the

can I hop in with a little bit about the health system Beyond kind of the vi

health system Beyond kind of the vi point so you know just broadly General

point so you know just broadly General Catalyst invest in healthcare I'm a life

Catalyst invest in healthcare I'm a life sciences specialist but learned from my

sciences specialist but learned from my colleagues in healthc care and some of

colleagues in healthc care and some of the things that we're seeing in AI for

the things that we're seeing in AI for health care and health delivery are

health care and health delivery are things like viz AI which help enhance

things like viz AI which help enhance efficacy or quality so can I enable a

efficacy or quality so can I enable a doctor to treat a patient faster which

doctor to treat a patient faster which drives a better outcome or to lower an

drives a better outcome or to lower an error rate in things like Radiology or

error rate in things like Radiology or pathology so that's one overall category

pathology so that's one overall category or to find sepsis based on synthesizing

or to find sepsis based on synthesizing various data and run an alert everyone

various data and run an alert everyone hates epic sepsis alert it's been around

hates epic sepsis alert it's been around forever it's not AI drives people crazy

forever it's not AI drives people crazy but can you actually drive better

but can you actually drive better quality with these kinds of alerting

quality with these kinds of alerting that's one category is how can you use

that's one category is how can you use AI to drive better efficacy then there's

AI to drive better efficacy then there's better throughput or lower cost

better throughput or lower cost investors get seduced by this but it's

investors get seduced by this but it's not what health systems are solving for

not what health systems are solving for right so like better throughput is nice

right so like better throughput is nice um you know there are various ways you

um you know there are various ways you know in Radiology or pathology that you

know in Radiology or pathology that you can use AI to move faster but what

can use AI to move faster but what people really get excited about is

people really get excited about is driving patient impact differential

driving patient impact differential outcomes and then the last aspect is is

outcomes and then the last aspect is is taking cost and friction out of the

taking cost and friction out of the system so doctors don't like doing notes

system so doctors don't like doing notes imagine some of you guys are doctors in

imagine some of you guys are doctors in the audience um and also when you think

the audience um and also when you think about the overall costs in the Health

about the overall costs in the Health Care System you're paying a lot of

Care System you're paying a lot of people for things like healthcare bling

people for things like healthcare bling coding or nurse scheduling right there's

coding or nurse scheduling right there's a big Workforce shortage you're actually

a big Workforce shortage you're actually paying someone a lot of money to

paying someone a lot of money to schedule and reschedule nurses so there

schedule and reschedule nurses so there are places where AI can slot in to

are places where AI can slot in to relieve administrative burden which

relieve administrative burden which frees up money for research freeze up

frees up money for research freeze up money for patient care and so things

money for patient care and so things like that are as simple as nurse

like that are as simple as nurse scheduling cure as a company in our

scheduling cure as a company in our portfolio that has that among really

portfolio that has that among really broads set of offerings things like

broads set of offerings things like helping patients prepare for their

helping patients prepare for their appointments and giving nurses leverage

appointments and giving nurses leverage in that um and things like Healthcare

in that um and things like Healthcare billing and coding which are very manual

billing and coding which are very manual and labor intensive um you know in an

and labor intensive um you know in an Ideal World we could remove that as a

Ideal World we could remove that as a job category but given that it is a

job category but given that it is a category how can you make it high

category how can you make it high quality and faster using AA so that's

quality and faster using AA so that's kind of how I think about it like within

kind of how I think about it like within the broader Health Care delivery side is

the broader Health Care delivery side is most exciting appeals to my heart

most exciting appeals to my heart driving better patient

driving better patient outcomes important is better throughput

outcomes important is better throughput and then indirect effect but really good

and then indirect effect but really good for the health system are where can you

for the health system are where can you take things that are manual and labor

take things that are manual and labor intensive and transform them to be less

intensive and transform them to be less labor intensive and more comp compute

labor intensive and more comp compute enabled thank you and I thought we could

enabled thank you and I thought we could turn to Ben to discuss Healthcare

turn to Ben to discuss Healthcare applications on the insurer side yeah

applications on the insurer side yeah absolutely uh so I think one of the main

absolutely uh so I think one of the main points I'd like to make at an overview

points I'd like to make at an overview level is that in as many of you know in

level is that in as many of you know in the health insurance space insurers do a

the health insurance space insurers do a lot more than just providing insurance

lot more than just providing insurance so I think I think of insurance business

so I think I think of insurance business models a lot as relating to kind of

models a lot as relating to kind of developing a collection or a basket of

developing a collection or a basket of different rationing policies or

different rationing policies or different care provision policies that

different care provision policies that kind of

kind of get kind of care delivered to consumers

get kind of care delivered to consumers in a way that the insurer feels is kind

in a way that the insurer feels is kind of efficient uh and effective for them

of efficient uh and effective for them and of course insur private insurers do

and of course insur private insurers do this across a range of markets Medicare

this across a range of markets Medicare Advantage Medicare Part D the Affordable

Advantage Medicare Part D the Affordable Care Act exchanges and large employer

Care Act exchanges and large employer markets uh I think you know there's

markets uh I think you know there's insurers have always used data as a core

insurers have always used data as a core part of their businesses uh but I think

part of their businesses uh but I think that in recent years data has become

that in recent years data has become even more core part of what insures are

even more core part of what insures are building as kind of a business and

building as kind of a business and building as assets and together with

building as assets and together with that together with those data assets um

that together with those data assets um complimentary expertise in data

complimentary expertise in data analytics uh I think something that's

analytics uh I think something that's quite important here is that you know

quite important here is that you know for insurers to gain competitive

for insurers to gain competitive advantages in this space uh they're

advantages in this space uh they're really working hard to develop

really working hard to develop proprietary data sets and develop

proprietary data sets and develop proprietary data sets that um really are

proprietary data sets that um really are kind of span the whole vertical supp

kind of span the whole vertical supp supply chain of healthcare not just you

supply chain of healthcare not just you know we're looking to try and kind of

know we're looking to try and kind of rate risk of certain customers but all

rate risk of certain customers but all sorts of details on metadata relating to

sorts of details on metadata relating to the interactions between Physicians and

the interactions between Physicians and patients metadata related to claims

patients metadata related to claims denial and prior

denial and prior authorization uh things like that and so

authorization uh things like that and so I think you know just some of the

I think you know just some of the functions I think of insurers leveraging

functions I think of insurers leveraging these data and techniques to work on

these data and techniques to work on include you know prior

include you know prior authorization uh for drugs and for

authorization uh for drugs and for services

services uh how do we Implement claims denials in

uh how do we Implement claims denials in a way that's uh consistent with our

a way that's uh consistent with our policies that's kind of how insurers

policies that's kind of how insurers would describe it uh how do we Implement

would describe it uh how do we Implement value based care policies that also May

value based care policies that also May rely on data related to uh electronic

rely on data related to uh electronic health records uh that they're uh

health records uh that they're uh bringing in uh and then on the insurance

bringing in uh and then on the insurance side how do we find customers on the

side how do we find customers on the employer side and split their risk in a

employer side and split their risk in a way that kind of allows us to be

way that kind of allows us to be profitable more recently I think data uh

profitable more recently I think data uh one thing insurers have been using data

one thing insurers have been using data for us to highlight providers uh that

for us to highlight providers uh that they can look to purchase uh and I think

they can look to purchase uh and I think that more broadly uh in this space

that more broadly uh in this space vertical integration between insurers

vertical integration between insurers and other businesses um such as Pharmacy

and other businesses um such as Pharmacy benefit managers or

benefit managers or providers uh has been obviously a very

providers uh has been obviously a very strong Trend and I think that you know

strong Trend and I think that you know when we've thought about these things

when we've thought about these things we've thought about these things in

we've thought about these things in terms of the vertical economics of

terms of the vertical economics of pricing and models uh on kind of the

pricing and models uh on kind of the more core economic side

more core economic side but I think that one of the reasons

but I think that one of the reasons insurers are really you know looking to

insurers are really you know looking to you know work to to to to vertically

you know work to to to to vertically integrate with these kinds of fir with

integrate with these kinds of fir with these kinds of firms is they're looking

these kinds of firms is they're looking to assemble these large data sets across

to assemble these large data sets across the whole supply chain to use to gain a

the whole supply chain to use to gain a competitive advantage in their

competitive advantage in their businesses um it's all stopped

businesses um it's all stopped there thank you and Elena I thought you

there thank you and Elena I thought you could give us the overview and examples

could give us the overview and examples in the Pharma and Life Sciences space

in the Pharma and Life Sciences space just a quick PLL how many people in the

just a quick PLL how many people in the audience are interested in health

audience are interested in health systems and insurers as your primary

systems and insurers as your primary interest area how many people live in

interest area how many people live in payer provider

payer provider World High hands okay that's either

World High hands okay that's either you're very shy or not that many how

you're very shy or not that many how many people live in life sciences

many people live in life sciences World okay and then most of you won't

World okay and then most of you won't vote

vote um this is like worse than us voter

um this is like worse than us voter turnout okay um we're just trying to

turnout okay um we're just trying to gauge I'm just trying to gauge here how

gauge I'm just trying to gauge here how much interest there is in life science

much interest there is in life science but I'll take that as a little bit um so

but I'll take that as a little bit um so if you think about life sciences as

if you think about life sciences as Therapeutics Diagnostics research tools

Therapeutics Diagnostics research tools and then software and service that

and then software and service that enables that there are opportunities

enables that there are opportunities across the Spectrum on the drug

across the Spectrum on the drug Discovery or therapeutic side it can

Discovery or therapeutic side it can apply to everything from your target

apply to everything from your target Discovery so like what biology are you

Discovery so like what biology are you going after or what biology matters to

going after or what biology matters to how you make your chemical matter so

how you make your chemical matter so whether it's a small molecule or an

whether it's a small molecule or an antibody understanding where am I trying

antibody understanding where am I trying to bind to so understanding the target

to bind to so understanding the target protein or what chemical am I going to

protein or what chemical am I going to make whether it's chemical or protein

make whether it's chemical or protein what structure am I making you can use

what structure am I making you can use AI in all of

AI in all of that um so there's different ways to

that um so there's different ways to leverage AI depending on what your

leverage AI depending on what your problem is but it's highly relevant up

problem is but it's highly relevant up and down the value chain of drug

and down the value chain of drug Discovery the same thing in Diagnostics

Discovery the same thing in Diagnostics and it's a little bit more concrete so

and it's a little bit more concrete so back to sepsis are you mining the EMR or

back to sepsis are you mining the EMR or aggregating other sources of data to

aggregating other sources of data to provide an early warning system and an

provide an early warning system and an early intervention in cancer can you do

early intervention in cancer can you do this difficult signal to noise detection

this difficult signal to noise detection problem of finding rare circulating

problem of finding rare circulating tumor uh DNA or other forms of cancer

tumor uh DNA or other forms of cancer and distinguishing that from normal

and distinguishing that from normal healthy DNA in the blood um so in

healthy DNA in the blood um so in Diagnostics it's about ident identifying

Diagnostics it's about ident identifying something and surfacing that for

something and surfacing that for intervention and then in research it's

intervention and then in research it's actually really cool something that

actually really cool something that happened here at

happened here at Stanford um by a fellow named Renee

Stanford um by a fellow named Renee Casey is he started a company called

Casey is he started a company called metal Loop that does endtoend enablement

metal Loop that does endtoend enablement of research so everything from helping

of research so everything from helping you write your Grant to data collection

you write your Grant to data collection so pulling data out of the EMR playing

so pulling data out of the EMR playing data out of wearable so how do you get

data out of wearable so how do you get all of this multimodal data to data

all of this multimodal data to data analysis to writing your paper at the

analysis to writing your paper at the other end so basically for each of these

other end so basically for each of these areas you can take Ai and apply it to

areas you can take Ai and apply it to every step in the value chain you can't

every step in the value chain you can't have it do it for you but you can get a

have it do it for you but you can get a starting draft for your Grant you can

starting draft for your Grant you can gather multimodal data more easily you

gather multimodal data more easily you can analyze it more easily and you can

can analyze it more easily and you can put your results out more easily and

put your results out more easily and that's maybe like the most concrete

that's maybe like the most concrete example of the same concept whether it's

example of the same concept whether it's Therapeutics diagnostics enablement for

Therapeutics diagnostics enablement for research is take whatever value chain

research is take whatever value chain you had before before and think about

you had before before and think about where can AI solve a problem that

where can AI solve a problem that couldn't be solved before or help me do

couldn't be solved before or help me do this better faster cheaper one other

this better faster cheaper one other example on research I'll mention is um

example on research I'll mention is um screening for study subjects right like

screening for study subjects right like how can you find study subjects faster

how can you find study subjects faster to power a study could be hugely

to power a study could be hugely valuable saves time right and and gets a

valuable saves time right and and gets a study completed sooner so if you can use

study completed sooner so if you can use AI to you know find those patients

AI to you know find those patients sometimes needles in the hay stack

sometimes needles in the hay stack it's much better than waiting for the

it's much better than waiting for the principal investigator to have that

principal investigator to have that person show up in their clinic or their

person show up in their clinic or their Ed it's a huge deal because right now

Ed it's a huge deal because right now nurses have to go through electronic

nurses have to go through electronic medical records surfacing hey I think

medical records surfacing hey I think this patient could be a good fit for

this patient could be a good fit for your trial and then the PI says maybe

your trial and then the PI says maybe I'm going to enroll them um but if you

I'm going to enroll them um but if you have something like viai they can pull

have something like viai they can pull through scans and in real time say hey

through scans and in real time say hey this patient could be a relevant subject

this patient could be a relevant subject for this intervention especially for

for this intervention especially for things where you need to identify and

things where you need to identify and enroll patients quickly for sort of

enroll patients quickly for sort of urgent or critical studies it's a huge

urgent or critical studies it's a huge challenge for how do you find the

challenge for how do you find the patient to offer them that intervention

patient to offer them that intervention and usually involves having a nurse

and usually involves having a nurse sitting right there waiting for a

sitting right there waiting for a patient to happen um and then like

patient to happen um and then like operationally for a place like Stanford

operationally for a place like Stanford anywhere else the question is are you

anywhere else the question is are you going to fund or staff a clinical

going to fund or staff a clinical research nurse at all times where you're

research nurse at all times where you're taking risk on that person's salary

taking risk on that person's salary while hoping to have a Pharma partner

while hoping to have a Pharma partner sponsor it that's kind of apart um but

sponsor it that's kind of apart um but if you have aib based tools that can

if you have aib based tools that can enable patient identification then you

enable patient identification then you can bring in a nurse when needed without

can bring in a nurse when needed without having to take that risk on Staffing and

having to take that risk on Staffing and it means you can do a lot more trials

it means you can do a lot more trials without requiring an external sponsor or

without requiring an external sponsor or without requiring you to Bear the burden

without requiring you to Bear the burden of that cost that's just like a good

of that cost that's just like a good example of like one kind of like it

example of like one kind of like it sounds like a simple problem but really

sounds like a simple problem but really hard problem of how do you go from this

hard problem of how do you go from this patient could benefit from a trial to

patient could benefit from a trial to actually giving them access to that

actually giving them access to that trial

trial with that very helpful background I

with that very helpful background I wanted to have each panelist go through

wanted to have each panelist go through and give their take on the state of um

and give their take on the state of um Healthcare AI competition today and so

Healthcare AI competition today and so the question is are we seeing uh the

the question is are we seeing uh the amount of competition we would expect in

amount of competition we would expect in a competitive Marketplace and if not

a competitive Marketplace and if not what are the barriers to competition and

what are the barriers to competition and to make sure we're hearing from everyone

to make sure we're hearing from everyone I'll start with zad who we haven't heard

I'll start with zad who we haven't heard from yet um thanks so I thought um

from yet um thanks so I thought um instead of directly answering the

instead of directly answering the question I'd tell you a story that

question I'd tell you a story that speaks to the state of the market so a

speaks to the state of the market so a few years ago with some colleagues we

few years ago with some colleagues we started evaluating a family of

started evaluating a family of algorithms that um is sold to a number

algorithms that um is sold to a number of healthare systems around the country

of healthare systems around the country so just to give you a sense of um the

so just to give you a sense of um the the problem these algorithms are solving

the problem these algorithms are solving you've got a population of patients you

you've got a population of patients you know some of them are going to get sick

know some of them are going to get sick if you knew which one of them was going

if you knew which one of them was going to get sick you would do a lot of things

to get sick you would do a lot of things to help them prevent exacerbations of

to help them prevent exacerbations of chronic conditions flare ups things like

chronic conditions flare ups things like that today but you can't do that for

that today but you can't do that for everyone so algorithms are perfect for

everyone so algorithms are perfect for this job you find the people who are

this job you find the people who are going to get sick and you help them

going to get sick and you help them today so it's a great AI application and

today so it's a great AI application and there are a number of large companies

there are a number of large companies that are selling software to do this to

that are selling software to do this to healthcare system so this is at the

healthcare system so this is at the intersection of kind of clinical and

intersection of kind of clinical and insurance and um and and optimization

insurance and um and and optimization and what we found was that these

and what we found was that these algorithms which are screening um

algorithms which are screening um between 100 and 200 million people per

between 100 and 200 million people per year for this you know extra help with

year for this you know extra help with your chronic conditions um contained

your chronic conditions um contained this very large amount of racial bias um

this very large amount of racial bias um so they were putting healthier white

so they were putting healthier white patients ahead in line for extra help um

patients ahead in line for extra help um ahead of sicker black patients um and

ahead of sicker black patients um and and this was not known it was like this

and this was not known it was like this was not an efficient situation like

was not an efficient situation like after the study came out uh after a lot

after the study came out uh after a lot of senators sent letters to companies

of senators sent letters to companies and Hospital systems and things like

and Hospital systems and things like that people stopped using this we worked

that people stopped using this we worked with the team at the company to fix the

with the team at the company to fix the problem in the this is not like a design

problem in the this is not like a design thing this is just an error um but it

thing this is just an error um but it was not known and this was like hundreds

was not known and this was like hundreds of millions of patients per year being

of millions of patients per year being affected by this fairly large scale

affected by this fairly large scale racial bias now why why did this happen

racial bias now why why did this happen why were we the ones who um figured it

why were we the ones who um figured it out well there's a very simple answer

out well there's a very simple answer which is that we had access to the data

which is that we had access to the data that we needed to basically um do this

that we needed to basically um do this analysis and show that black patients

analysis and show that black patients were being discriminated against by this

were being discriminated against by this algorithm um and and I think that's a

algorithm um and and I think that's a microcosm of a much larger problem in

microcosm of a much larger problem in this area which is that the consumer is

this area which is that the consumer is not given the data that he or she needs

not given the data that he or she needs to make the decision about the algorithm

to make the decision about the algorithm that um that they're that they're

that um that they're that they're purchasing and I think that that is a

purchasing and I think that that is a huge problem and it and it limits the

huge problem and it and it limits the competitive this SM the competitiveness

competitive this SM the competitiveness of this Market it it prevents the market

of this Market it it prevents the market from being aimed in the right direction

from being aimed in the right direction and helping uh patients and helping the

and helping uh patients and helping the healthare

healthare system um Ben do you w to go next sure

system um Ben do you w to go next sure yeah um I'll be brief and I'll just talk

yeah um I'll be brief and I'll just talk a little bit about the role of data in

a little bit about the role of data in in in Insurance markets um you know I

in in Insurance markets um you know I think that data are just uh and this

think that data are just uh and this kind of a glomeration of data as an

kind of a glomeration of data as an asset it's kind of one ingredient that

asset it's kind of one ingredient that has kind of led to economies of scale

has kind of led to economies of scale for insurers both across kind of

for insurers both across kind of horizontal markets where insurers are

horizontal markets where insurers are merging with other insurers as well as

merging with other insurers as well as vertical markets where insurers are

vertical markets where insurers are merging with other parts of the um other

merging with other parts of the um other parts of the supply chain and I think

parts of the supply chain and I think that in Health policy we've seen this in

that in Health policy we've seen this in the past with things like accountable

the past with things like accountable care organizations but I think in in

care organizations but I think in in merger policy now both on the economic

merger policy now both on the economic side but also on on the legal side

side but also on on the legal side there's this tension where you know if

there's this tension where you know if if a merger leads to tangible

if a merger leads to tangible efficiencies or it's a vertical merger

efficiencies or it's a vertical merger and there's some presumed

and there's some presumed efficiencies uh we might kind of

efficiencies uh we might kind of historically de facto let this merger go

historically de facto let this merger go ahead when in fact when you have data

ahead when in fact when you have data and data analytics is kind of a central

and data analytics is kind of a central asset that's a part of this merger

asset that's a part of this merger it can lead to a number of asymmetries

it can lead to a number of asymmetries and a number of competitive advantages

and a number of competitive advantages where you know it's at some point it's

where you know it's at some point it's not obvious that these kind of

not obvious that these kind of increasing economies of scale are worth

increasing economies of scale are worth the trade-off with with Market power and

the trade-off with with Market power and I think that you know the new merger

I think that you know the new merger guidelines for example try to speak to

guidelines for example try to speak to these issues uh but I think that there

these issues uh but I think that there needs to be both kind of more

needs to be both kind of more development on the economic side but

development on the economic side but also uh on the on the law side on the

also uh on the on the law side on the legal side

legal side here Allison yeah just building on some

here Allison yeah just building on some of the points that the other panelists

of the points that the other panelists have made I think in the healthcare

have made I think in the healthcare space data is such an asset and as Ben

space data is such an asset and as Ben pointed out it has continued to grow as

pointed out it has continued to grow as an asset as more and more folks have

an asset as more and more folks have started to use artificial intelligence

started to use artificial intelligence but the other thing I think that's

but the other thing I think that's pretty unique about Healthcare that we

pretty unique about Healthcare that we have to think about with data is privacy

have to think about with data is privacy and that's something that uh also limits

and that's something that uh also limits the sharing of data the identification

the sharing of data the identification of healthcare data is really complicated

of healthcare data is really complicated if I know your doctor and your

if I know your doctor and your Healthcare conditions and your ZIP code

Healthcare conditions and your ZIP code I can probably figure out who you are

I can probably figure out who you are using social media or other publicly

using social media or other publicly available data and so I think there's

available data and so I think there's this kind of holding on to data because

this kind of holding on to data because we can get value from it but there's

we can get value from it but there's also holding on to data because we're

also holding on to data because we're worried about the risks to patients and

worried about the risks to patients and I think those are two things we we have

I think those are two things we we have to balance as we start to think about

to balance as we start to think about how we can make data more available to

how we can make data more available to promote competition in this space

Elena yeah um the panelists have covered it really well so far so just add maybe

it really well so far so just add maybe two small thoughts one is when we invest

two small thoughts one is when we invest in companies we take a look from the

in companies we take a look from the responsible Innovation perspective and

responsible Innovation perspective and so what we mean by that is we think

so what we mean by that is we think about intended and unintended

about intended and unintended consequences if this company scales what

consequences if this company scales what are the incentives that they will face

are the incentives that they will face and how will their behavior change as a

and how will their behavior change as a result of that so during the pandemic

result of that so during the pandemic you know a company called cerebral was

you know a company called cerebral was doing uh remote um psychiatric care

doing uh remote um psychiatric care which is really good on its face right

which is really good on its face right expand patient access to mental health

expand patient access to mental health the challenge is as a fee for service

the challenge is as a fee for service entity they were highly incentivized to

entity they were highly incentivized to churn patients and write lots of

churn patients and write lots of prescriptions that's just an example of

prescriptions that's just an example of how do you think about the intended and

how do you think about the intended and unintended consequences the intention

unintended consequences the intention was to expand patient access and address

was to expand patient access and address unmet medical need and the unintended

unmet medical need and the unintended consequence was an incentive to

consequence was an incentive to overprescribe or prescribe quickly

overprescribe or prescribe quickly without care um and so I think the same

without care um and so I think the same can be applied when you think about

can be applied when you think about businesses right what is the market

businesses right what is the market structure and how will this Market

structure and how will this Market structure incentivize or disincentivize

structure incentivize or disincentivize competition and then there are really

competition and then there are really smart people here uh whose job it is to

smart people here uh whose job it is to guide policy um or Sue uh to block uh

guide policy um or Sue uh to block uh unintended consequences and so we just

unintended consequences and so we just try to invest in companies where we

try to invest in companies where we think that their intention and their

think that their intention and their risk factors align with our mission um

risk factors align with our mission um and then count on the smart folks around

and then count on the smart folks around the table here to uh create the right

the table here to uh create the right Market structures for them to respond

Market structures for them to respond to and then I'll just say overall I

to and then I'll just say overall I think in the kind of the the viz space

think in the kind of the the viz space applying AI to images to detect things

applying AI to images to detect things and get patients are added to the care

and get patients are added to the care they need um Elena's firm uh has a

they need um Elena's firm uh has a portfolio comp that does very similar

portfolio comp that does very similar things so right on this panel you have

things so right on this panel you have two competitors and right here at

two competitors and right here at Stanford our third competitor uh was

Stanford our third competitor uh was created by a physician at Stanford so

created by a physician at Stanford so you know Ju Just Right Here in in this

you know Ju Just Right Here in in this this area we've got three Market

this area we've got three Market competitors so um but I do share what

competitors so um but I do share what what Andrew Ang mentioned at the end of

what Andrew Ang mentioned at the end of his speech the concern about data access

his speech the concern about data access so as um less so with medical images

so as um less so with medical images right I think we have Pathways to get

right I think we have Pathways to get access to that data but as you get into

access to that data but as you get into generative AI tools applied to health

generative AI tools applied to health record data you you do have powerful

record data you you do have powerful players that um hold that data and

players that um hold that data and there's there's other tools outside of I

there's there's other tools outside of I think anit trust you've got like cares

think anit trust you've got like cares act information blocking rules right so

act information blocking rules right so so long as they continue to play nicely

so long as they continue to play nicely and they allow Innovative companies to

and they allow Innovative companies to access that data then um then I'm not

access that data then um then I'm not concerned but if if that changes then I

concerned but if if that changes then I think it's something to keep an eye on

think it's something to keep an eye on yeah and I think our whole field owes a

yeah and I think our whole field owes a debt to the Regulators on making patient

debt to the Regulators on making patient data portable right like as a patient I

data portable right like as a patient I want to be able to take my data to

want to be able to take my data to wherever I seek care and then to

wherever I seek care and then to patients for then consenting their data

patients for then consenting their data into research um so I think that's a

into research um so I think that's a really good counterbalance regulation

really good counterbalance regulation that prevents the epics of the world

that prevents the epics of the world from keeping all of their data in a

from keeping all of their data in a Walled Garden or a health system saying

Walled Garden or a health system saying you can't take your data and switch to

you can't take your data and switch to another Health System there's actually

another Health System there's actually some really great regulation that lets

some really great regulation that lets you take your data with you and the next

you take your data with you and the next step could be can I take my algorithms

step could be can I take my algorithms with me so if someone develops an

with me so if someone develops an algorithm that's benefiting me as a

algorithm that's benefiting me as a patient can I Port that over just like

patient can I Port that over just like Regulators have said I'm allowed to Port

Regulators have said I'm allowed to Port my phone number can I take my health

my phone number can I take my health algorithms with me wherever I go I just

algorithms with me wherever I go I just want to point out one strange thing

want to point out one strange thing about this Market which and David you

about this Market which and David you brought up you know there are these

brought up you know there are these three companies doing similar things if

three companies doing similar things if you're hesitating among three cars you

you're hesitating among three cars you can go test drive the cars and you can

can go test drive the cars and you can make your decision one of the things

make your decision one of the things that's powered all of the advances in AI

that's powered all of the advances in AI over the past few decades is the ability

over the past few decades is the ability to see here are three algorithms let's

to see here are three algorithms let's just see how they're performing on this

just see how they're performing on this one gold standard data set and let's be

one gold standard data set and let's be able to compare their performance that's

able to compare their performance that's just what the algorithm is supposed to

just what the algorithm is supposed to be doing outputting a number on this

be doing outputting a number on this data set let's just see which one does

data set let's just see which one does it best and we can't do that with AI in

it best and we can't do that with AI in healthcare because there's so few ways

healthcare because there's so few ways for everyone to access these kinds of

for everyone to access these kinds of evaluations data set so it's taking the

evaluations data set so it's taking the The Supercharged fuel of AI out of

The Supercharged fuel of AI out of healthcare because we can't have access

healthcare because we can't have access to data where we can do these kinds of

to data where we can do these kinds of comparisons yeah and we kind of talked

comparisons yeah and we kind of talked about this a little bit in in the prep

about this a little bit in in the prep for this but the um I agree and like you

for this but the um I agree and like you can't necessarily like line up up like

can't necessarily like line up up like cars and say like this one has this much

cars and say like this one has this much horsepower and that one has that much

horsepower and that one has that much horsepower but um algorithm performance

horsepower but um algorithm performance I don't think is the fundamental issue

I don't think is the fundamental issue right it's like does it actually help

right it's like does it actually help patients um improve outcomes does it

patients um improve outcomes does it help reduce costs uh so at some point

help reduce costs uh so at some point you know the algorithm is is good enough

you know the algorithm is is good enough at doing what it's supposed to do but

at doing what it's supposed to do but the the the less exciting part of the AI

the the the less exciting part of the AI is the whole workflow tool on Healthcare

is the whole workflow tool on Healthcare the care coordination piece right

the care coordination piece right because if you have an amazing algorithm

because if you have an amazing algorithm but it goes nowhere and a clinician

but it goes nowhere and a clinician doesn't act on it a patient's outcomes

doesn't act on it a patient's outcomes aren't going to improve so um I I agree

aren't going to improve so um I I agree with you it would be wonderful to be

with you it would be wonderful to be able to kind of Benchmark and test and

able to kind of Benchmark and test and say Here's the actual positive

say Here's the actual positive predictive value of algorithm a b and c

predictive value of algorithm a b and c um but what I'm more interested in is

um but what I'm more interested in is like those real world studies done by

like those real world studies done by academic medical centers where they say

academic medical centers where they say hey before this algorithm was

hey before this algorithm was implemented what was our uh door to

implemented what was our uh door to needle time for stroke right after it my

needle time for stroke right after it my goodness we reduce it by 30 minutes and

goodness we reduce it by 30 minutes and that is like literally saving people's

that is like literally saving people's lives sending people home who are no

lives sending people home who are no longer disabled reducing costs for

longer disabled reducing costs for healthare right for for payers as well

healthare right for for payers as well because you're not supporting someone

because you're not supporting someone who may be disabled for the rest of

who may be disabled for the rest of their life so I I agree with you but I

their life so I I agree with you but I think it's more than just like straight

think it's more than just like straight up algorithm performance it would be

up algorithm performance it would be like it would be like buying a car based

like it would be like buying a car based upon horsepower and then realizing it's

upon horsepower and then realizing it's a two-seater and I can't get my kids to

a two-seater and I can't get my kids to soccer practice right and I think that

soccer practice right and I think that you know when you have a standard of

you know when you have a standard of evidence where you want to show impact

evidence where you want to show impact on patient outcomes if you want to run a

on patient outcomes if you want to run a randomized trial I think that would be

randomized trial I think that would be great like that but that seems like a

great like that but that seems like a whole other level of claims and evidence

whole other level of claims and evidence let's first figure out if the are

let's first figure out if the are working the way they're supposed to yeah

working the way they're supposed to yeah and some of that I will say like is is

and some of that I will say like is is um for The Med device algorithms right

um for The Med device algorithms right maybe I don't think the Epic sepsis ALG

maybe I don't think the Epic sepsis ALG algorithm for example was cleared as a

algorithm for example was cleared as a medical device but for um the algorithms

medical device but for um the algorithms that detect stroke for example they they

that detect stroke for example they they have to be cleared so there is some

have to be cleared so there is some floor level that FDA requires and also

floor level that FDA requires and also addresses some of the issues around bias

addresses some of the issues around bias right making sure that the data isn't

right making sure that the data isn't just from like one um you know one one

just from like one um you know one one hospital or or one type of patient and

hospital or or one type of patient and so um you have a diverse data set and it

so um you have a diverse data set and it performs at least at some minimum level

performs at least at some minimum level so you don't alert fatigue

so you don't alert fatigue Physicians I mean I think not to be

Physicians I mean I think not to be sitting in the Venture Corner over here

sitting in the Venture Corner over here but just to Echo what you're saying viz

but just to Echo what you're saying viz AI uh and AI doc the company we're

AI uh and AI doc the company we're invested in uh and any other AI player a

invested in uh and any other AI player a health system can say hey before I buy

health system can say hey before I buy it I want to run a pilot and I'm going

it I want to run a pilot and I'm going to run a pilot with all your competitors

to run a pilot with all your competitors let's see how it goes um and I think

let's see how it goes um and I think where viai and where AI doc have really

where viai and where AI doc have really differentiated is to your point in the

differentiated is to your point in the workflow like I met viai three four

workflow like I met viai three four years ago as an investor so my numbers

years ago as an investor so my numbers are totally out of date but I remember

are totally out of date but I remember then it was in like 400 Health Systems

then it was in like 400 Health Systems and had taken off like wildflyer not

and had taken off like wildflyer not because of the algorithm performance per

because of the algorithm performance per se but because of the alerting system to

se but because of the alerting system to Physicians it let them solve a problem

Physicians it let them solve a problem and the same thing with AI do it fits

and the same thing with AI do it fits into the physician workflow there's

into the physician workflow there's another company whose name I won't name

another company whose name I won't name that went bankrupt which was funded by

that went bankrupt which was funded by one of my competitors and they went to

one of my competitors and they went to Market saying we're going to replace

Market saying we're going to replace Radiologists well guess what

Radiologists well guess what Radiologists buy the AI um and they want

Radiologists buy the AI um and they want to know like does this make my life

to know like does this make my life easier does this drive patients outcomes

easier does this drive patients outcomes and then maybe I still want a job

and then maybe I still want a job tomorrow um so they didn't buy that one

tomorrow um so they didn't buy that one um you know and to your point um this is

um you know and to your point um this is like an audience participation uh

like an audience participation uh opportunity didn't the FDA say say to

opportunity didn't the FDA say say to Epic hey your sepsis thing should be

Epic hey your sepsis thing should be regulated anyone remember that anybody

regulated anyone remember that anybody anybody um not to speak out of school

anybody um not to speak out of school but I'm almost certain they said hey

but I'm almost certain they said hey this needs to come under software as a

this needs to come under software as a medical device regulation um so you do

medical device regulation um so you do have like a good balance there of saying

have like a good balance there of saying like hey there does have to be some bar

like hey there does have to be some bar for does this work um before you make it

for does this work um before you make it available

available right I think we had some really

right I think we had some really interesting discussion there on the

interesting discussion there on the dimensions of competition

dimensions of competition um I want to turn back to one of the

um I want to turn back to one of the earlier themes which is data access as a

earlier themes which is data access as a bottleneck and how Market participants

bottleneck and how Market participants who don't have ownership of a lot of

who don't have ownership of a lot of data can access it and I think um zad

data can access it and I think um zad and David both have interesting stories

and David both have interesting stories here so start with zad can you tell us

here so start with zad can you tell us about what prompted you to found night

about what prompted you to found night Andale and dandelion which both seem to

Andale and dandelion which both seem to be about um giving wider access to data

be about um giving wider access to data that people and companies can use to run

that people and companies can use to run a

a yeah I think a huge problem with the way

yeah I think a huge problem with the way this space is currently structured is

this space is currently structured is that the data live in one administrative

that the data live in one administrative enclave and the people who know how to

enclave and the people who know how to do machine learning live in a completely

do machine learning live in a completely different administrative enclave and it

different administrative enclave and it is really hard to either get the data

is really hard to either get the data out or to get those people in and so um

out or to get those people in and so um Nightingale open science is a nonprofit

Nightingale open science is a nonprofit um and this was a solution to the

um and this was a solution to the problem of lots of Berkeley PhD students

problem of lots of Berkeley PhD students coming to me and saying like oh I know

coming to me and saying like oh I know how to do this cool Machine Vision thing

how to do this cool Machine Vision thing it seems like it could be really useful

it seems like it could be really useful in healthcare I'd like to get involved

in healthcare I'd like to get involved in one and what I would say is like

in one and what I would say is like great we'll get your criminal background

great we'll get your criminal background check and your fingerprints done and

check and your fingerprints done and then we'll add you to the IRB and the

then we'll add you to the IRB and the data use agreement and then in two years

data use agreement and then in two years you'll and by that time of course

you'll and by that time of course they're already like working at Facebook

they're already like working at Facebook doing adclick optimization um and not

doing adclick optimization um and not doing Healthcare so uh Nightingale is um

doing Healthcare so uh Nightingale is um philanthropically funded we raised a few

philanthropically funded we raised a few million dollars from I'd say Tech Quant

million dollars from I'd say Tech Quant forward philanthropies Eric Schmidt Ken

forward philanthropies Eric Schmidt Ken Griffin Gordon and Betty Moore and we

Griffin Gordon and Betty Moore and we use that to create data sets around

use that to create data sets around interesting unsolved medical problems we

interesting unsolved medical problems we de identify the data inside of the

de identify the data inside of the health system we bring it out put it on

health system we bring it out put it on our Cloud you can access it within 24

our Cloud you can access it within 24 hours after you sign the data use

hours after you sign the data use agreement it's Nightingale science.org

agreement it's Nightingale science.org um so that is a kind of one data set at

um so that is a kind of one data set at a time solution uh it's not a market

a time solution uh it's not a market solution and so dandelion is a

solution and so dandelion is a for-profit company um that I uh helped

for-profit company um that I uh helped co-found and dandelion partners with um

co-found and dandelion partners with um a a set of large non-academic Health

a a set of large non-academic Health Systems where we get access to um all of

Systems where we get access to um all of their data and are able to De identify

their data and are able to De identify it within their infrastructure bring it

it within their infrastructure bring it out onto our cloud and then um allow a

out onto our cloud and then um allow a carefully curated set of clinical

carefully curated set of clinical algorithm developers to get access to

algorithm developers to get access to those data to build clinical products so

those data to build clinical products so this is a way to get the market um

this is a way to get the market um working and pointed in the right

working and pointed in the right direction for algorithm development so

direction for algorithm development so that people aren't just comp competing

that people aren't just comp competing on access to data but are competing on

on access to data but are competing on algorithm

algorithm quality and that would definitely be

quality and that would definitely be hugely valuable right because getting

hugely valuable right because getting getting the data so um uh every time viz

getting the data so um uh every time viz for example or ad do or any other

for example or ad do or any other company in the space moves into another

company in the space moves into another type of data you've got to gather that

type of data you've got to gather that data um it's not just that getting the

data um it's not just that getting the data too it's also annotating the data

data too it's also annotating the data you have to need to know whether it's

you have to need to know whether it's positive or false you know not not a

positive or false you know not not a positive it's a negative um scan

positive it's a negative um scan uh so uh how did viz do it initially I

uh so uh how did viz do it initially I think was by doing research agreements

think was by doing research agreements right but that's expensive it's time

right but that's expensive it's time consuming like you said you have to go

consuming like you said you have to go through irbs um uh it's costly because

through irbs um uh it's costly because academic medical centers have to have uh

academic medical centers have to have uh research coordinators people who pull

research coordinators people who pull data right so um it it it takes quite a

data right so um it it it takes quite a bit of time from there I will say once

bit of time from there I will say once um you have a commercial product

um you have a commercial product typically in the United States you'll be

typically in the United States you'll be able to De identify data um there's a

able to De identify data um there's a there's a little there's interesting

there's a little there's interesting interesting wrinkle with an FDA cleared

interesting wrinkle with an FDA cleared algorithm is that currently FDA doesn't

algorithm is that currently FDA doesn't allow that algorithm to improve in real

allow that algorithm to improve in real time right like if you use I don't know

time right like if you use I don't know like Gmail for example and it starts to

like Gmail for example and it starts to realize every time you sign your emails

realize every time you sign your emails with best regards it starts to suggest

with best regards it starts to suggest best regards right it doesn't have to be

best regards right it doesn't have to be trained with your data outside of that

trained with your data outside of that but for um cleared algorithms FDA

but for um cleared algorithms FDA cleared algorithms you have to train in

cleared algorithms you have to train in a non clinical setting so so when we

a non clinical setting so so when we engage our customers um healthc Care

engage our customers um healthc Care Systems like Stanford or or others uh we

Systems like Stanford or or others uh we get we negotiate the right to De

get we negotiate the right to De identify that data and train the

identify that data and train the algorithms and it's for the benefit of

algorithms and it's for the benefit of the Health Care system because their

the Health Care system because their data is being used to improve the

data is being used to improve the algorithm it's for the benefit of of viz

algorithm it's for the benefit of of viz and patience because the algorithm gets

and patience because the algorithm gets gets better and then you've got that

gets better and then you've got that potentially depending on the rights you

potentially depending on the rights you have you have that that data for other

have you have that that data for other purposes um but if you go to a whole new

purposes um but if you go to a whole new type of data like ultrasound right you'd

type of data like ultrasound right you'd have to somehow source that data and if

have to somehow source that data and if there was a way to do that to partner

there was a way to do that to partner with a company like yours that would be

with a company like yours that would be hugely beneficial because it's not easy

hugely beneficial because it's not easy to find the

to find the data Ben you talked a little bit about

data Ben you talked a little bit about how large data owners might have a

how large data owners might have a competitive Advantage here um can you

competitive Advantage here um can you share a little bit more about that

share a little bit more about that thinking do we see large data owners

thinking do we see large data owners taking advantage of their data access to

taking advantage of their data access to to do Health Care AI applications or to

to do Health Care AI applications or to disadvantage rivals in any way so any

disadvantage rivals in any way so any general thoughts there yeah I mean I I

general thoughts there yeah I mean I I think the answer is very likely yes

think the answer is very likely yes because they have the they have the

because they have the they have the incentive to do so uh and in some ways

incentive to do so uh and in some ways the fact that the data are proprietary

the fact that the data are proprietary make it harder to to assess whether

make it harder to to assess whether that's that's actually happening um I

that's that's actually happening um I think one good example of this is the um

think one good example of this is the um the United Healthcare merger with uh

the United Healthcare merger with uh change Healthcare where I was a uh

change Healthcare where I was a uh expert witness for the for the doj and

expert witness for the for the doj and there you know United was acquiring

there you know United was acquiring change and one of the aspects of this

change and one of the aspects of this merger is United is gaining access as a

merger is United is gaining access as a broad parent company to this great meta

broad parent company to this great meta data set that has lots of information on

data set that has lots of information on kind of the secret sauce rationing

kind of the secret sauce rationing Technologies of many of their Rivals

Technologies of many of their Rivals rival insurers so you can kind of name

rival insurers so you can kind of name all the other uh insurers you want and

all the other uh insurers you want and as you know by being able to acquire

as you know by being able to acquire that data set and own that data they

that data set and own that data they gain a serious competitive advantage in

gain a serious competitive advantage in terms of being able in real time to

terms of being able in real time to assess and learn about the strategies of

assess and learn about the strategies of their Rivals right and it for example if

their Rivals right and it for example if that merger didn't happen and change was

that merger didn't happen and change was an independent company it could be that

an independent company it could be that they're going to develop products that

they're going to develop products that are more publicly available or

are more publicly available or purchasable by all the

purchasable by all the insurers uh who can then use that to run

insurers uh who can then use that to run analytics for example on deidentified

analytics for example on deidentified data so I think it's kind of an

data so I think it's kind of an insurance side vers version of what zad

insurance side vers version of what zad was talking about and what we were

was talking about and what we were talking about in the health space but I

talking about in the health space but I think there's a real just kind of

think there's a real just kind of important discussion about you know what

important discussion about you know what data sets do we want to live in a

data sets do we want to live in a proprietary space under under companies

proprietary space under under companies and what data sets do we want to at

and what data sets do we want to at least live in a firm that's kind of

least live in a firm that's kind of independent of kind of the companies are

independent of kind of the companies are using it versus live under kind of a

using it versus live under kind of a more like regulator

more like regulator purview and to that point I wanted to

purview and to that point I wanted to ask what are the alternatives to

ask what are the alternatives to corporate Mass in of data in this

corporate Mass in of data in this space um is this should we be thinking

space um is this should we be thinking about as Po from a policy perspective

about as Po from a policy perspective making some of this data publicly

making some of this data publicly available through a regulator and I know

available through a regulator and I know CMS already does some of that so I

CMS already does some of that so I thought maybe Allison could talk a

thought maybe Allison could talk a little bit about that yeah so we do a

little bit about that yeah so we do a lot of work to make CMS data available

lot of work to make CMS data available for research um but it's the way the had

for research um but it's the way the had talked about you have to come through a

talked about you have to come through a data use agreement you have to tell us

data use agreement you have to tell us what research you're doing and also does

what research you're doing and also does a lot of work to make data available for

a lot of work to make data available for research but I think uh the role that

research but I think uh the role that Regulators can play is more what Elena

Regulators can play is more what Elena was talking about earlier the work that

was talking about earlier the work that CMS and onc have done to say Can giving

CMS and onc have done to say Can giving patients the ability to share their data

patients the ability to share their data with whoever they want to share it with

with whoever they want to share it with if they want their data to be used to

if they want their data to be used to support AI development great if they

support AI development great if they want to change their doctor they can

want to change their doctor they can take their data with them so allowing

take their data with them so allowing the flow of data encouraging the flow of

the flow of data encouraging the flow of data developing standards in this space

data developing standards in this space so that you know if you're working with

so that you know if you're working with one data set and you want to work with a

one data set and you want to work with a different data set you're not completely

different data set you're not completely rethinking your code development and all

rethinking your code development and all of that I think there's a lot more that

of that I think there's a lot more that that's kind of the space that we can

that's kind of the space that we can play in and allowing The Innovation to

play in and allowing The Innovation to happen that zad is doing for example

happen that zad is doing for example with night and Gale to say how do we

with night and Gale to say how do we open this data up how do we make it more

open this data up how do we make it more available how do we think about really

available how do we think about really Innovative techniques for de

Innovative techniques for de identification but also

identification but also you know to the point I was making early

you know to the point I was making early about earlier about privacy even when

about earlier about privacy even when the data is deidentified you're still

the data is deidentified you're still coming into an enclave you're still

coming into an enclave you're still signing an agreement saying you're not

signing an agreement saying you're not going to re-identify an individual so I

going to re-identify an individual so I think there's there's work that

think there's there's work that Regulators are doing to make data

Regulators are doing to make data available to encourage the seamless

available to encourage the seamless exchange of data but also allowing

exchange of data but also allowing Innovation to happen in the

Innovation to happen in the marketplace and I think we are running

marketplace and I think we are running out of time um do we do we have five

out of time um do we do we have five more minutes or we have one more minute

more minutes or we have one more minute okay do we need to take audience

okay do we need to take audience questions

questions now want to

now want to St okay well well let's see if the

St okay well well let's see if the audience has any questions we should

audience has any questions we should open it I won't hog these amazing

open it I won't hog these amazing panelists

panelists anymore but if you don't have questions

anymore but if you don't have questions I have

I have more okay well any anyone have a

more okay well any anyone have a question okay I wanted to ask um we

question okay I wanted to ask um we talked about data access as a a

talked about data access as a a bottleneck to competition I wanted to

bottleneck to competition I wanted to see are there other bottlenecks in the

see are there other bottlenecks in the space we should be thinking

space we should be thinking about I I I mean I think one that is

about I I I mean I think one that is that's pretty clear is data is kind of

that's pretty clear is data is kind of bottlenecks in in labor and

bottlenecks in in labor and sophisticated labor so um you know data

sophisticated labor so um you know data data Engineers data analysts uh you know

data Engineers data analysts uh you know a lot of the biggest companies get

a lot of the biggest companies get access to a lot of the best data

access to a lot of the best data engineers and data analysts and you know

engineers and data analysts and you know their expertise is crucial in developing

their expertise is crucial in developing these tools and so I think that's a

these tools and so I think that's a potential

potential bottleneck I think there's a particular

bottleneck I think there's a particular kind of expertise that in these domains

kind of expertise that in these domains like health or law or um hiring or

like health or law or um hiring or wherever AI is being used you actually

wherever AI is being used you actually need people who are kind of bilingual

need people who are kind of bilingual who understand that domain but also

who understand that domain but also understand how to do all of the data

understand how to do all of the data stuff and I think our educational system

stuff and I think our educational system is not very good at producing those

is not very good at producing those kinds of people unfortunately so ways to

kinds of people unfortunately so ways to uh get more of those um bying people I

uh get more of those um bying people I think would be really useful I'm really

think would be really useful I'm really optimistic because of work like what

optimistic because of work like what Allison is doing to standardize data and

Allison is doing to standardize data and make it more open um to address the

make it more open um to address the bottleneck in terms of you know data

bottleneck in terms of you know data analysis data engineers and then with

analysis data engineers and then with Next Generation models you don't need as

Next Generation models you don't need as much data like the early days of viai or

much data like the early days of viai or AI do was all about the painstaking

AI do was all about the painstaking aggregation and then manual in some

aggregation and then manual in some cases annotation of data and so with

cases annotation of data and so with Next Generation models with llms with

Next Generation models with llms with generative AI you can have zero shot or

generative AI you can have zero shot or small shot algorithm development and so

small shot algorithm development and so data becomes less of a bottleneck to the

data becomes less of a bottleneck to the application layer it's the bottleneck to

application layer it's the bottleneck to the base layer models but as Healthcare

the base layer models but as Healthcare people I think like it's an incredibly

people I think like it's an incredibly exciting time for application layer

and do patients have this vast knowledge of their data and the data the algorithm

of their data and the data the algorithm and whether they can transfer it or when

and whether they can transfer it or when they change Insurance especially you

they change Insurance especially you know the older populations I think uh

know the older populations I think uh there's a lot of role for just

there's a lot of role for just disseminating that information maybe uh

disseminating that information maybe uh to the patients I'm not sure if that's a

to the patients I'm not sure if that's a yeah I agree I think it is something

yeah I agree I think it is something that there's there's starting to be

that there's there's starting to be awareness that patients have this

awareness that patients have this ownership of their data and their right

ownership of their data and their right the right to take their data with them

the right to take their data with them but it is something that we need to

but it is something that we need to continue to do education on Medicare for

continue to do education on Medicare for example has a a service that we call

example has a a service that we call Blue Buton 2.0 which allows Medicare

Blue Buton 2.0 which allows Medicare beneficiaries to direct their data to

beneficiaries to direct their data to any application or service that they

any application or service that they want to use we haven't seen a huge

want to use we haven't seen a huge uptake especially among the you know the

uptake especially among the you know the Medicare age population who maybe aren't

Medicare age population who maybe aren't as as digitally savvy but we are

as as digitally savvy but we are starting to see growth in those types of

starting to see growth in those types of programs as we do more education and as

programs as we do more education and as more applications come on board and do

more applications come on board and do Outreach to beneficiaries offering their

services I want to thank the panelists so much for very eliminating insights

so much for very eliminating insights today I wish we had more time to talk we

today I wish we had more time to talk we could talk all afternoon but we have a

could talk all afternoon but we have a lot of other interesting panels so thank

lot of other interesting panels so thank you so much please join me in giving

you so much please join me in giving them a big round of applause

you all right we're going to move on to our next uh panel of the day after we

our next uh panel of the day after we studied our Healthcare applications

studied our Healthcare applications we're going to study uh and learn more

we're going to study uh and learn more about uh chips and Hardware that can

about uh chips and Hardware that can make some of these applications run and

make some of these applications run and we're very uh pleased to Welcome to the

we're very uh pleased to Welcome to the stage we have our uh two uh uh folks

stage we have our uh two uh uh folks that will be engaging in a fire side

that will be engaging in a fire side chat um first we have um Paula blizzard

chat um first we have um Paula blizzard who is the senior uh Assistant Attorney

who is the senior uh Assistant Attorney General in the antitrust section of the

General in the antitrust section of the California AG's office and she will

California AG's office and she will having be having this discussion with

having be having this discussion with the Victor Pang who is president at AMD

the Victor Pang who is president at AMD and has very deep knowledge of the chip

and has very deep knowledge of the chip industry so please let me welcome uh

industry so please let me welcome uh Victor and uh thank him for uh talking

Victor and uh thank him for uh talking to us about uh hardware and uh chips in

to us about uh hardware and uh chips in this space and uh then we will be joined

this space and uh then we will be joined by the rest of our panel uh discussion

by the rest of our panel uh discussion um in about 15 minutes so uh thank you

um in about 15 minutes so uh thank you Victor thank

Victor thank you thank you very much thank you to

you thank you very much thank you to Stanford and the US Department of

Stanford and the US Department of Justice for organizing this um as uh was

Justice for organizing this um as uh was mentioned I'm with the California

mentioned I'm with the California government um we are obviously very

government um we are obviously very focused and invested and excited about

focused and invested and excited about everything thing going on uh with AI

everything thing going on uh with AI here in

here in California um Victor Pang is the

California um Victor Pang is the president of AMD as was said he has um

president of AMD as was said he has um 40 years of experience in Innovative

40 years of experience in Innovative hardware and Chip development and he's

hardware and Chip development and he's responsible for the company's AI

responsible for the company's AI strategy um thank you for coming Victor

strategy um thank you for coming Victor thank you very much it's great to be

thank you very much it's great to be here um so hardware and chips have again

here um so hardware and chips have again a bit taken Center Stage they are

a bit taken Center Stage they are crucial to all the things uh that make

crucial to all the things uh that make these large language models run um and

these large language models run um and yet they are at the bottom of the stack

yet they are at the bottom of the stack those of you who think about Stacks

those of you who think about Stacks right the software at chat gpts at the

right the software at chat gpts at the top the chips at the bottom okay most of

top the chips at the bottom okay most of us will never touch one we'll never see

us will never touch one we'll never see one but they're super important so

one but they're super important so Victor help us out here give us a little

Victor help us out here give us a little bit of an overview what types of chips

bit of an overview what types of chips are important what do they do give just

are important what do they do give just a little primer on some of the acronyms

a little primer on some of the acronyms would be great yeah so for for some of

would be great yeah so for for some of you who um already can follow stochastic

you who um already can follow stochastic gradian descent the bear with me because

gradian descent the bear with me because it'll be a little bit more basic um you

it'll be a little bit more basic um you know look I think most people have heard

know look I think most people have heard about CPUs and certainly AMD you know

about CPUs and certainly AMD you know and and others have been uh in decades

and and others have been uh in decades in that business but you can think of

in that business but you can think of the CPU as a general jack of all trades

the CPU as a general jack of all trades right it could run all kinds of

right it could run all kinds of applications and software um everything

applications and software um everything from things that are more consumer like

from things that are more consumer like uh like on your your notebook or your

uh like on your your notebook or your laptop top you know web browsers your

laptop top you know web browsers your personal productivity video streaming or

personal productivity video streaming or in the server in the data centers

in the server in the data centers running transaction processing or other

running transaction processing or other kinds of more Enterprise kind of

kinds of more Enterprise kind of applications right so now you hear about

applications right so now you hear about gpus and gpus stands for graphics

gpus and gpus stands for graphics Processing Unit um so and you you sort

Processing Unit um so and you you sort of say well you know it's for AI so why

of say well you know it's for AI so why is it you know call Graphics Well turns

is it you know call Graphics Well turns out that you know graphics and AI um

out that you know graphics and AI um from a workload perspective has

from a workload perspective has similarities it's very um linear Al very

similarities it's very um linear Al very intensive um lots of Matrix

intensive um lots of Matrix multiplications it's also extremely

multiplications it's also extremely parallel um lots of data movement lots

parallel um lots of data movement lots of high bandwidth so it turns out

of high bandwidth so it turns out initially um before AI really exploded

initially um before AI really exploded on the scene researchers were doing a

on the scene researchers were doing a lot of uh development on Graphics gpus

lot of uh development on Graphics gpus right um and just confusing a little bit

right um and just confusing a little bit further like today we have uh two

further like today we have uh two different families of Graphics one

different families of Graphics one that's still for mainly for graphics

that's still for mainly for graphics processing gaming and and and

processing gaming and and and visualization and

visualization and but then we have another family that's

but then we have another family that's concentrated really on performing really

concentrated really on performing really large um deployment scale out extremely

large um deployment scale out extremely large and also in high performance

large and also in high performance Computing and this they're both qu gpus

Computing and this they're both qu gpus but in any event um then then you you've

but in any event um then then you you've probably heard of other kind of AI

probably heard of other kind of AI accelerat like tpus um you know lots of

accelerat like tpus um you know lots of different startups that have many

different startups that have many different architectures they're all

different architectures they're all broadly speaking a form of accelerator

broadly speaking a form of accelerator for AI applications right so um the the

for AI applications right so um the the other thing I would say say from

other thing I would say say from semiconductors to click down one level

semiconductors to click down one level two is that when you have these massive

two is that when you have these massive deployments of you know tens to hundreds

deployments of you know tens to hundreds of thousands and people are talking

of thousands and people are talking about you know millions of gpus um at

about you know millions of gpus um at that point too you also really need to

that point too you also really need to have a really good networking

have a really good networking architecture because you're taking these

architecture because you're taking these models or you're taking know large

models or you're taking know large workloads and you're actually dividing

workloads and you're actually dividing it across the entire uh data center you

it across the entire uh data center you know multiple Ikeas size buildings large

know multiple Ikeas size buildings large and you know your Bic um often times at

and you know your Bic um often times at that point is is more the network as

that point is is more the network as opposed to the compute on on giving node

opposed to the compute on on giving node um so there is a lot of complexity and

um so there is a lot of complexity and and maybe the the last thing I'll say

and maybe the the last thing I'll say even though there's a good deal of

even though there's a good deal of summarization going on in Andrew's talk

summarization going on in Andrew's talk which makes sense you know he just

which makes sense you know he just showed at the very bottom of the layer

showed at the very bottom of the layer as you say but even at that you know I

as you say but even at that you know I talked a lot about Hardware but there's

talked a lot about Hardware but there's a tremendous amount of software and

a tremendous amount of software and firmware you know embedded software so

firmware you know embedded software so to speak that makes that work otherwise

to speak that makes that work otherwise you really can't get any useful work out

you really can't get any useful work out of it so um these are systems that are

of it so um these are systems that are both hardware and software and it's all

both hardware and software and it's all really important to actually get the

really important to actually get the endtoend uh performance that you need

endtoend uh performance that you need yeah super interesting Okay so we've got

yeah super interesting Okay so we've got these Ikeas sized buildings full of

these Ikeas sized buildings full of billions upon billions of these little

billions upon billions of these little chips and they're all somehow being

chips and they're all somehow being coordinated with some software and some

coordinated with some software and some firmware um and then it goes on up to

firmware um and then it goes on up to ultimately sort of at some point get to

ultimately sort of at some point get to chat GPT and we're putting in little

chat GPT and we're putting in little little queries but let me ask you do do

little queries but let me ask you do do all of these things have to be

all of these things have to be integrated really well to work together

integrated really well to work together and what I'm thinking of is do we need

and what I'm thinking of is do we need one company that vertically integrates

one company that vertically integrates from the chip all the way up or are

from the chip all the way up or are there certain sort of demarcation

there certain sort of demarcation points yeah you know I I think um you

points yeah you know I I think um you know going back to the fact that you

know going back to the fact that you know AI is such a full stack you know

know AI is such a full stack you know and Andrew talked about the application

and Andrew talked about the application Level and and other middle layers and

Level and and other middle layers and you know the whole model was just one

you know the whole model was just one layer closest to the foundational

layer closest to the foundational Hardware layer um and as you said if you

Hardware layer um and as you said if you really you know poke down one more

really you know poke down one more there's actually more detail I guess

there's actually more detail I guess what I'd say is uh this is what makes

what I'd say is uh this is what makes both AI very interesting but also very

both AI very interesting but also very complex is you know we actually do have

complex is you know we actually do have to be aware certainly of the models but

to be aware certainly of the models but even higher up in in the stack and to

even higher up in in the stack and to make everything perform and we have to

make everything perform and we have to be aware to to Really devise a really

be aware to to Really devise a really good GPU and by the way CPUs are also

good GPU and by the way CPUs are also needed um it's it's GPU rich but you

needed um it's it's GPU rich but you also need CPUs and then you need the

also need CPUs and then you need the networking as I mentioned um

networking as I mentioned um so I guess what I say to have a

so I guess what I say to have a performant overall an system there has

performant overall an system there has to be at minimum kind of what I would

to be at minimum kind of what I would say co-optimization or or co-

say co-optimization or or co- understanding so it could be delivered

understanding so it could be delivered fully integrated but it should not only

fully integrated but it should not only be delivered that way um because you

be delivered that way um because you know basically you know the you could

know basically you know the you could have Innovation at every level and

have Innovation at every level and Andrew talked about that like it's

Andrew talked about that like it's tremendous activity in multiple levels

tremendous activity in multiple levels um and that's a really good thing

um and that's a really good thing because that drives you know better uh

because that drives you know better uh delivery uh lower cost higher

delivery uh lower cost higher performance enabling more powerful

performance enabling more powerful models enabling and performance just to

models enabling and performance just to make that concrete for everybody the

make that concrete for everybody the speed at which you get a response from

speed at which you get a response from your prompt that that's inference

your prompt that that's inference performance right and the amount of time

performance right and the amount of time and cost it takes you to train a new

and cost it takes you to train a new model maybe to do um you know bit more

model maybe to do um you know bit more accurate Imaging analysis of MRIs you

accurate Imaging analysis of MRIs you know that's a training kind of uh

know that's a training kind of uh performance level so it does matter

performance level so it does matter right um and so um the other thing I

right um and so um the other thing I would say is so you really should

would say is so you really should allow Innovation at every point and I

allow Innovation at every point and I think you know we've got tremendous

think you know we've got tremendous expertise in in AMD and and other

expertise in in AMD and and other companies but I I think it's hard to

companies but I I think it's hard to imagine that you know you would have the

imagine that you know you would have the best at every point in time at every

best at every point in time at every level of that stack so I do think it's

level of that stack so I do think it's important from an innovation and

important from an innovation and competitive and just forward progress to

competitive and just forward progress to sort of enable multiple players and uh

sort of enable multiple players and uh Andrew also mentioned open source you

Andrew also mentioned open source you know we we are well we're a big company

know we we are well we're a big company and we're for profit uh we embrace the

and we're for profit uh we embrace the open source from a perspective of of our

open source from a perspective of of our software stack where we open source a

software stack where we open source a tremendous amount of our software stack

tremendous amount of our software stack now there's still elements in areas

now there's still elements in areas where it's it's really super core to our

where it's it's really super core to our business that we don't necessarily do

business that we don't necessarily do that but I I just want to share that an

that but I I just want to share that an example of like not every big company

example of like not every big company that's that has a significant um you

that's that has a significant um you know stake in AI is taking the approach

know stake in AI is taking the approach sure everything is fully proprietary our

sure everything is fully proprietary our strategy in fact is General to be even

strategy in fact is General to be even if it's not open source the partner with

if it's not open source the partner with multiple other uh players like we're

multiple other uh players like we're doing that on the networking space right

doing that on the networking space right so we're enabling you know an ecosystem

so we're enabling you know an ecosystem of multiple players and we're enabling

of multiple players and we're enabling in that sense choice and I also think it

in that sense choice and I also think it drives you know Innovation more rapidly

drives you know Innovation more rapidly all right well I am all for more

all right well I am all for more Innovation but um it sounds like the

Innovation but um it sounds like the systems are massively huge physically as

systems are massively huge physically as well as um complicated and it needs to

well as um complicated and it needs to all work together really well so when

all work together really well so when you type in your prompt you get a fast

you type in your prompt you get a fast answer that is also accurate and has

answer that is also accurate and has been uh the system has been trained well

been uh the system has been trained well and yet to get the Innovation it has to

and yet to get the Innovation it has to be open at points so give us some more

be open at points so give us some more examples of how where in this stock it

examples of how where in this stock it can be more open um and uh I'll just

can be more open um and uh I'll just throw the last question to you a bit so

throw the last question to you a bit so should the government do that or this

should the government do that or this are the companies going to do it how's

are the companies going to do it how's AMD doing it and how are you looking at

AMD doing it and how are you looking at this problem well you know again we we

this problem well you know again we we are taking a strategy of um you know

are taking a strategy of um you know being enabling an ecosystem and working

being enabling an ecosystem and working with uh ecosystems and other partners

with uh ecosystems and other partners and you know it's not all you know in

and you know it's not all you know in the in the purest form open source but

the in the purest form open source but it is enabling multiple players and also

it is enabling multiple players and also interoperability I think you mentioned

interoperability I think you mentioned that before I didn't directly address

that before I didn't directly address that that's the other thing is even if

that that's the other thing is even if you do um feel like you have a turnkey

you do um feel like you have a turnkey solution and some customers want a

solution and some customers want a turnkey solution many customers have

turnkey solution many customers have their own special needs there really

their own special needs there really isn't a one-size fits-all and the other

isn't a one-size fits-all and the other thing is that this is a very very fast

thing is that this is a very very fast moving uh technology in Marketplace at

moving uh technology in Marketplace at every level of that full stack so I

every level of that full stack so I think you know I would say you know

think you know I would say you know enabling that um in the government's

enabling that um in the government's role what I would say is like you know

role what I would say is like you know you see how the Market's performing if

you see how the Market's performing if it's doing its right thing and

it's doing its right thing and competition is um you know it's a fair

competition is um you know it's a fair and Level Playing Field like you know we

and Level Playing Field like you know we we have leadership products you know we

we have leadership products you know we believe we're going to win that um and

believe we're going to win that um and we're taking a different approach a more

we're taking a different approach a more open approach and you know bringing

open approach and you know bringing other uh Co companies along with us um

other uh Co companies along with us um and we get really good feedback quite

and we get really good feedback quite frankly from a lot of our customers on

frankly from a lot of our customers on that um the key thing there though is

that um the key thing there though is that if it for whatever reason becomes

that if it for whatever reason becomes less of a Level Playing Field then I

less of a Level Playing Field then I think there is certainly a role for

think there is certainly a role for government and in one sense it's not

government and in one sense it's not different than what government has

different than what government has always done but I think the the

always done but I think the the additional challenge is it is very

additional challenge is it is very complex and it's changing very rapidly

complex and it's changing very rapidly and you know I completely agree with

and you know I completely agree with Andrew and it's interesting I've

Andrew and it's interesting I've interactions with Andrew but usually not

interactions with Andrew but usually not in this kind of

in this kind of context so so we never we didn't share

context so so we never we didn't share notes but you know his thing about uh

notes but you know his thing about uh the fact of you know look at what's

the fact of you know look at what's appropriate for the application and what

appropriate for the application and what the challenge the use case if you will

the challenge the use case if you will and then also uh Percy's comment about

and then also uh Percy's comment about marginal value versus marginal risk um

marginal value versus marginal risk um in fact internally we we've adopted

in fact internally we we've adopted because we have a responsible AI as well

because we have a responsible AI as well uh as you know within as a function

uh as you know within as a function within uh AMD and we look at both what

within uh AMD and we look at both what we're doing outbound technology we also

we're doing outbound technology we also look at how we use technology internally

look at how we use technology internally we we use the N for uh framework for

we we use the N for uh framework for risk identification and same thing it's

risk identification and same thing it's like you know there's certain risks if

like you know there's certain risks if you're talking about Medical Imaging in

you're talking about Medical Imaging in people's Health outcomes that that you

people's Health outcomes that that you know requires higher rigor perhaps

know requires higher rigor perhaps potentially higher guard rails but at

potentially higher guard rails but at the same time you want to enable data

the same time you want to enable data right so it doesn't mean everything is

right so it doesn't mean everything is you know it's a sort of um control and

you know it's a sort of um control and other areas where you know the outcome

other areas where you know the outcome of a recommending you know you didn't

of a recommending you know you didn't get my movie preference just right like

get my movie preference just right like you know okay you don't need to regulate

you know okay you don't need to regulate that too heavily so I think being uh

that too heavily so I think being uh commerate with the risk factor and also

commerate with the risk factor and also you know measuring uh marginal gain and

you know measuring uh marginal gain and marginal risk those are important things

marginal risk those are important things and just seeing if the marketplace is a

and just seeing if the marketplace is a level or not excellent thank you okay so

level or not excellent thank you okay so complicated for government as well

complicated for government as well because we're going to have to do it on

because we're going to have to do it on a risk basis one size is one regulation

a risk basis one size is one regulation will not fit all and whether you like

will not fit all and whether you like the Barbie or Oppenheimer or whether you

the Barbie or Oppenheimer or whether you need to get and MRI are probably very

need to get and MRI are probably very different questions um the time for our

different questions um the time for our little fireside chat is over so now I'm

little fireside chat is over so now I'm going to invite um or Jennifer is going

going to invite um or Jennifer is going to invite the rest of our panelists up

to invite the rest of our panelists up but we will continue the discussion um

but we will continue the discussion um about hardware and

about hardware and chips thank you Victor welcome we're

chips thank you Victor welcome we're going to do a little uh reshuffling here

going to do a little uh reshuffling here uh as as politics takes the seat of

uh as as politics takes the seat of moderator and we're going to welcome up

moderator and we're going to welcome up our panelists as I said all their full

our panelists as I said all their full full bios are uh available online um but

full bios are uh available online um but I will I will at least welcome them with

I will I will at least welcome them with their titles our our first panelist is

their titles our our first panelist is Chris wolf he's the global head of AI um

Chris wolf he's the global head of AI um and Advanced Services uh by by VMware at

and Advanced Services uh by by VMware at broadcom uh we have missar amiman

broadcom uh we have missar amiman founder and CEO of occi labs we have

founder and CEO of occi labs we have Alex Gainer is a colle at the FTC he's

Alex Gainer is a colle at the FTC he's the chief uh deputy chief technologist

the chief uh deputy chief technologist there and we do have a uh participant by

there and we do have a uh participant by by Zoom uh blanch uh saur uh board she's

by Zoom uh blanch uh saur uh board she's general counsel and Secretary of the

general counsel and Secretary of the board at mrol AI so she will be joining

board at mrol AI so she will be joining us by video um and the rest of our

us by video um and the rest of our panelists are up here on stage so I will

panelists are up here on stage so I will leave it to Paula thank

leave it to Paula thank you thank you very much welcome to my uh

you thank you very much welcome to my uh distinguished panel welcome to blunch as

distinguished panel welcome to blunch as well hello um

well hello um uh we were just chatting um with Victor

uh we were just chatting um with Victor Pang of AMD about hardware and chips and

Pang of AMD about hardware and chips and now we're going to get some other

now we're going to get some other perspectives we're going to broaden it

perspectives we're going to broaden it um to see what other players and

um to see what other players and stakeholders in the market um are

stakeholders in the market um are thinking about and what I'd like to do

thinking about and what I'd like to do is just go to um each panelist and give

is just go to um each panelist and give them a minute to tell you about the

them a minute to tell you about the position that they are coming from um

position that they are coming from um their initial thoughts a little bit

their initial thoughts a little bit about their company and what role um

about their company and what role um they're playing in the development vment

they're playing in the development vment of hardware and chips so Chris um Chris

of hardware and chips so Chris um Chris wolf Global head of AI and advanced

wolf Global head of AI and advanced systems VMware by broadcom give us some

systems VMware by broadcom give us some thoughts okay thanks for the thanks for

thoughts okay thanks for the thanks for the opportunity and I would say I think

the opportunity and I would say I think most of you are probably familiar with

most of you are probably familiar with with broadcom leader and semiconductors

with broadcom leader and semiconductors and also a lot of data center software

and also a lot of data center software from security to Mainframe to

from security to Mainframe to virtualization I'd say one thing that's

virtualization I'd say one thing that's been unique about my role is that we in

been unique about my role is that we in addition to Leading the product

addition to Leading the product development side for AI we also operate

development side for AI we also operate the internal AI services and that's

the internal AI services and that's driven our product strategy so some some

driven our product strategy so some some examples we run a chat service for

examples we run a chat service for product support and uh we're seeing just

product support and uh we're seeing just incredible efficiencies and accuracy uh

incredible efficiencies and accuracy uh from that particular service and this is

from that particular service and this is using an open source model but the meta

using an open source model but the meta point was when we started running our

point was when we started running our own internal AI Services we realized

own internal AI Services we realized that it would be difficult to take a uh

that it would be difficult to take a uh single beted on one one vendor or

single beted on one one vendor or technology provider we wanted to take a

technology provider we wanted to take a platform approach that would allow us to

platform approach that would allow us to very quickly on board uh new AI services

very quickly on board uh new AI services and new AI models at the speed of

and new AI models at the speed of software our Customer Support Service as

software our Customer Support Service as an example has had its foundation model

an example has had its foundation model changed three times in the last nine

changed three times in the last nine months uh because there's been better

months uh because there's been better technology that's come along and that's

technology that's come along and that's what's led us to what we call Private AI

what's led us to what we call Private AI which is about bringing the AI model to

which is about bringing the AI model to where your data is uh you can you can

where your data is uh you can you can see a lot of efficiencies that way you

see a lot of efficiencies that way you can maintain privacy and control of your

can maintain privacy and control of your data while still gaining the benefits of

data while still gaining the benefits of AI and we've seen this really resonate

AI and we've seen this really resonate with our customer base and that's where

with our customer base and that's where we uh are pushing some new solutions

we uh are pushing some new solutions that we have that

that we have that it's not just a A broadcom or VMware

it's not just a A broadcom or VMware perspective we see this as really

perspective we see this as really industry and there's lots of ways to

industry and there's lots of ways to gain these similar benefits but it

gain these similar benefits but it doesn't have to be a trade-off I'd say

doesn't have to be a trade-off I'd say is my first point and the second is you

is my first point and the second is you you can gain these benefits with

you can gain these benefits with probably less gpus than you think like

probably less gpus than you think like Percy mentioned retrieval augmented

Percy mentioned retrieval augmented generation this morning one of our

generation this morning one of our Flagship Services is running on 4 gpus

Flagship Services is running on 4 gpus and uh this is you know when you bring

and uh this is you know when you bring in Rag and inference you can see really

in Rag and inference you can see really incredible results with the smaller

incredible results with the smaller investment and often times more

investment and often times more predictable costs and a lower carbon

predictable costs and a lower carbon footprint so there's there's a lot out

footprint so there's there's a lot out there I mean there's lots of it's it's

there I mean there's lots of it's it's not about you know AI being good or bad

not about you know AI being good or bad is about how are you approach an AI and

is about how are you approach an AI and what's best for the organization but

what's best for the organization but there's a lot of choice that you can you

there's a lot of choice that you can you can take on to do things differently

can take on to do things differently great thank you okay forget the Ikeas

great thank you okay forget the Ikeas sized rooms we can do it with four gpus

sized rooms we can do it with four gpus which are probably in the size of your

which are probably in the size of your phone um uh let me jump over um to maer

phone um uh let me jump over um to maer Meo founder and CEO of oy Labs formerly

Meo founder and CEO of oy Labs formerly at bit fusion and VMware what are your

at bit fusion and VMware what are your opening thoughts uh yeah Maz M at a Labs

opening thoughts uh yeah Maz M at a Labs um so our mission is to quantify the

um so our mission is to quantify the world um so as you can imagine AI is

world um so as you can imagine AI is sort of everywhere that's sort of the

sort of everywhere that's sort of the buzzword and I think we're at a very

buzzword and I think we're at a very interesting time where we can actually

interesting time where we can actually drive efficiencies in sectors that don't

drive efficiencies in sectors that don't see a lot of efficiency so one of the

see a lot of efficiency so one of the first areas that we're looking at is in

first areas that we're looking at is in the charitable nonprofit

the charitable nonprofit space um so I don't know if you know

space um so I don't know if you know this is a huge space about2 trillion go

this is a huge space about2 trillion go through uh us-based nonprofits every

through uh us-based nonprofits every year $2 trillion um of the $2 trillion

year $2 trillion um of the $2 trillion $500 billion goes to charitable

$500 billion goes to charitable nonprofits so uh places that are not

nonprofits so uh places that are not hospitals or insurance companies um and

hospitals or insurance companies um and of the $500 billion $180 billion only go

of the $500 billion $180 billion only go to program expenses so that's 36% and

to program expenses so that's 36% and who knows how much of that is actually

who knows how much of that is actually effective maybe 10 or 20% and so this is

effective maybe 10 or 20% and so this is U one area where um you know we've

U one area where um you know we've engineered a vision system that allows

engineered a vision system that allows us to um basically capture human

us to um basically capture human development indicators food production

development indicators food production Water Production um literacy rate um how

Water Production um literacy rate um how much milk is being produced by a goat

much milk is being produced by a goat farm for example and we think that

farm for example and we think that creating this feedback loop will make

creating this feedback loop will make charal giving way more impactful um so

charal giving way more impactful um so that's kind of what we're looking on and

that's kind of what we're looking on and kind of answer your question about uh

kind of answer your question about uh competition um so we we have like a lot

competition um so we we have like a lot of uh thoughts on this I would kind of

of uh thoughts on this I would kind of crystallize it to three one is that AI

crystallize it to three one is that AI is not a differentiator and there's very

is not a differentiator and there's very little value capture from AI

little value capture from AI functionality intelligence or uh machine

functionality intelligence or uh machine learning or data science um and similar

learning or data science um and similar to that is that software is also not a

to that is that software is also not a differentiator it's sort of an expected

differentiator it's sort of an expected outcome the marginal cost of creating a

outcome the marginal cost of creating a new line of code is approaching zero and

new line of code is approaching zero and the marginal cost of creating software

the marginal cost of creating software that creates software is also

that creates software is also approaching zero so what do we take from

approaching zero so what do we take from that um we are not an AI company as a

that um we are not an AI company as a rule uh but we will use uh Ai and

rule uh but we will use uh Ai and software uh to accelerate our

software uh to accelerate our engineering so agente AI we accelerate

engineering so agente AI we accelerate our engineering our testing our

our engineering our testing our robustness in our in our services and of

robustness in our in our services and of course uh we uh use code generators uh

course uh we uh use code generators uh as fast as possible because as a startup

as fast as possible because as a startup we're trying to find the truth as as

we're trying to find the truth as as soon as possible right uh you know

soon as possible right uh you know satisfy a human need and see if you can

satisfy a human need and see if you can extract value from the market um so

extract value from the market um so that's a second rule third rule is deep

that's a second rule third rule is deep Tech is probably uh can be accelerated

Tech is probably uh can be accelerated that I think this is where a lot of

that I think this is where a lot of innovation and opportunity is so um do

innovation and opportunity is so um do we need Better Health Care yes do we

we need Better Health Care yes do we need to increase our lifespan yes do we

need to increase our lifespan yes do we need uh access to better food yes do we

need uh access to better food yes do we need to clean our environment yes these

need to clean our environment yes these are all deep Tech problems that could be

are all deep Tech problems that could be accelerated with AI and I think that's

accelerated with AI and I think that's where a lot of the Innovation and

where a lot of the Innovation and opportunities are excellent thank you

opportunities are excellent thank you very much um ai ai is not a

very much um ai ai is not a differentiator I thought we were here to

differentiator I thought we were here to talk about how it was um and um but we

talk about how it was um and um but we got in marginal costs so I'm certain

got in marginal costs so I'm certain that Susan is Happy um let me jump

that Susan is Happy um let me jump across the ocean all the way to France

across the ocean all the way to France um blanch thank you so much for joining

um blanch thank you so much for joining us and staying up late a bit um please

us and staying up late a bit um please give us a little bit of background about

give us a little bit of background about mrol and some opening thoughts and thank

mrol and some opening thoughts and thank you very much for having me um so yeah

you very much for having me um so yeah I'm I'm General councel and Secretary of

I'm I'm General councel and Secretary of the board as at mistal which is m a

the board as at mistal which is m a French company that has been founded a

French company that has been founded a year ago and that ALS also gives a sense

year ago and that ALS also gives a sense of how fast this environment is evolving

of how fast this environment is evolving uh because in a year we've been able to

uh because in a year we've been able to build and launch several uh Frontier

build and launch several uh Frontier models we've been able to uh do two

models we've been able to uh do two series uh to fund

series uh to fund fundraisings and uh and and a lot more

fundraisings and uh and and a lot more including launching a platform and and

including launching a platform and and building several Partnerships with some

building several Partnerships with some of which are public like Microsoft

of which are public like Microsoft iws or or snowflake so um that's going

iws or or snowflake so um that's going very fast we're only 55 right now so two

very fast we're only 55 right now so two months ago we were 35 and and that also

months ago we were 35 and and that also gives I think a sense of what it is to

gives I think a sense of what it is to be a startup in AI that has the ambition

be a startup in AI that has the ambition to compete against Giants because

to compete against Giants because because it is what it is uh so the

because it is what it is uh so the funders came from those giants because

funders came from those giants because they came from meta and and from Deep

they came from meta and and from Deep Mind and they wanted to to build

Mind and they wanted to to build something a bit different uh in Europe

something a bit different uh in Europe uh with the talents that we have also uh

uh with the talents that we have also uh in Europe uh and with a lot of funding

in Europe uh and with a lot of funding including a lot of American Funding

including a lot of American Funding which which is reality so I think it's

which which is reality so I think it's also important to to note that uh so um

also important to to note that uh so um what is interesting about M trial is uh

what is interesting about M trial is uh we have no connection to uh to any

we have no connection to uh to any pre-existing actors in the tech industry

pre-existing actors in the tech industry uh so we of course have certain tickets

uh so we of course have certain tickets in the fundraising of some of these

in the fundraising of some of these actors and we have some Partnerships

actors and we have some Partnerships with them but we have no connection in

with them but we have no connection in the sense that we are we are a pure

the sense that we are we are a pure player we are not integrated on on the

player we are not integrated on on the stack and we are in the middle of the

stack and we are in the middle of the value chain uh building uh building

value chain uh building uh building Frontier models and that makes our

Frontier models and that makes our position super interesting because of

position super interesting because of course we have a lot of work to create

course we have a lot of work to create the good relationships to get everything

the good relationships to get everything including the compute that is the topic

including the compute that is the topic of today to get everything to to make it

of today to get everything to to make it work uh but we are also in a very good

work uh but we are also in a very good position to be available everywhere

position to be available everywhere where the customers want us to be and to

where the customers want us to be and to be deployed the way they want us to be

be deployed the way they want us to be deployed so we Deploy on premise we

deployed so we Deploy on premise we Deploy on clouds we Deploy on our

Deploy on clouds we Deploy on our platforms and and many other deployments

platforms and and many other deployments could be could be considered so that's

could be could be considered so that's uh that's also a key differentiator that

uh that's also a key differentiator that makes our position interesting from a

makes our position interesting from a competition standpoint um and of course

competition standpoint um and of course and I heard a lot earlier about

and I heard a lot earlier about commitment Victor's commitment to open

commitment Victor's commitment to open source and I think we can can say the

source and I think we can can say the same some some of the models we we

same some some of the models we we launched have been on aachi too and we

launched have been on aachi too and we will keep doing that and uh we launched

will keep doing that and uh we launched yesterday cod cod estral a model

yesterday cod cod estral a model dedicated for code that was under our

dedicated for code that was under our first uh draft um mixed license meaning

first uh draft um mixed license meaning that the weights are open but uh it's

that the weights are open but uh it's for testing so so that's also an

for testing so so that's also an interesting experiment for for this

interesting experiment for for this model uh and we've had already a lot of

model uh and we've had already a lot of adoption so that's that's what we do so

adoption so that's that's what we do so um yeah I think this is what I wanted to

um yeah I think this is what I wanted to say about M maybe to finish on this

say about M maybe to finish on this introduction uh we've had already a lot

introduction uh we've had already a lot of interactions with Regulators in

of interactions with Regulators in Europe uh so I think uh Mo probably most

Europe uh so I think uh Mo probably most of you have but because those Regulators

of you have but because those Regulators are very active these days and but we

are very active these days and but we have been at the Forefront of that

have been at the Forefront of that experiment too and uh the challenges uh

experiment too and uh the challenges uh this environment is going very fast so

this environment is going very fast so the challenges are only going to grow

the challenges are only going to grow very fast but Regulators have to are

very fast but Regulators have to are considering acting fast or reacting fast

considering acting fast or reacting fast in a world where it's very difficult to

in a world where it's very difficult to anticipate the side effects of of those

anticipate the side effects of of those actions and reactions so uh I completely

actions and reactions so uh I completely sympathize with with the uh the effort

sympathize with with the uh the effort that is really super important and and

that is really super important and and I'm willing to contribute to how we can

I'm willing to contribute to how we can think of a an action that makes a lot of

think of a an action that makes a lot of sense for for keeping this industry very

sense for for keeping this industry very open thanks a lot great thank thank you

open thanks a lot great thank thank you love the perspective congratulations on

love the perspective congratulations on getting a product out yesterday um uh

getting a product out yesterday um uh okay let me turn briefly to Alex Alec

okay let me turn briefly to Alex Alec Gainer is a deputy chief technologist at

Gainer is a deputy chief technologist at the FTC the FTC and many other agencies

the FTC the FTC and many other agencies are bringing in more and more tech

are bringing in more and more tech people to help with these um these

people to help with these um these Market challenges but um maybe you can

Market challenges but um maybe you can give us a little introduction from how

give us a little introduction from how you see all this playing out sure good

you see all this playing out sure good morning so the role of uh the team I

morning so the role of uh the team I work on at the FDC is to provide

work on at the FDC is to provide technical expertise to the commission

technical expertise to the commission and to the attorneys we work with on uh

and to the attorneys we work with on uh how these Technologies show up in the

how these Technologies show up in the agency's enforcement work both in

agency's enforcement work both in consumer protection as well as in

consumer protection as well as in competition and so I should say that uh

competition and so I should say that uh you know these views are my own and

you know these views are my own and don't necessarily represent any of any

don't necessarily represent any of any individual commissioner the commission

individual commissioner the commission as a whole but uh a top level uh way I

as a whole but uh a top level uh way I Orient myself in this space is that

Orient myself in this space is that there is no single question of what is

there is no single question of what is competition in AI I've got at least

competition in AI I've got at least three that I think are you know top

three that I think are you know top level questions one is what are the

level questions one is what are the competitive Dynamics in companies

competitive Dynamics in companies building models themselves particularly

building models themselves particularly building foundational or Frontier models

building foundational or Frontier models so that that's one market uh or one

so that that's one market uh or one question of competitive Dynamics you

question of competitive Dynamics you might have uh a second question you

might have uh a second question you might have is how does AI

might have is how does AI impact uh the competitive Dynamics in

impact uh the competitive Dynamics in markets particularly in the supply chain

markets particularly in the supply chain for AI the chips cloud computing orchest

for AI the chips cloud computing orchest ation tools all the other software and

ation tools all the other software and Hardware you need to actually build and

Hardware you need to actually build and deploy these models that's a second

deploy these models that's a second question you might ask about the

question you might ask about the competitive Dynamics and a third

competitive Dynamics and a third question you might ask

question you might ask is uh how will AI impact competition in

is uh how will AI impact competition in markets it's deployed in whether that's

markets it's deployed in whether that's Healthcare or Finance uh or any other

Healthcare or Finance uh or any other Market that you might find uses uh for

Market that you might find uses uh for these models um and these questions

these models um and these questions don't uh this the state of competition

don't uh this the state of competition in any one of these questions doesn't

in any one of these questions doesn't necessarily have the same answer and

necessarily have the same answer and there are interventions you might

there are interventions you might consider that will improve competition

consider that will improve competition in one but might not impact another so I

in one but might not impact another so I think uh I will do my best to as I'm

think uh I will do my best to as I'm saying you know this has potential to be

saying you know this has potential to be good this has potential to be bad to be

good this has potential to be bad to be specific where where do I think it is

specific where where do I think it is the potential to be good or bad or you

the potential to be good or bad or you know show promise or any of those other

know show promise or any of those other things because I think I think it's very

things because I think I think it's very easy to talk past each other when

easy to talk past each other when something has potential advantages in

something has potential advantages in one area or disadvantages in another

one area or disadvantages in another there doesn't have to be a single answer

all right sounds like uh you've got your work cut out for you um Lots going on um

work cut out for you um Lots going on um now let me BR just bring it back to

now let me BR just bring it back to Victor and you can uh react a little bit

Victor and you can uh react a little bit to all the other panelists and or give

to all the other panelists and or give us a little more on amd's thoughts on

us a little more on amd's thoughts on yeah let me first do that because we

yeah let me first do that because we only kind of talked about CPS and GPS

only kind of talked about CPS and GPS but um you know AMD actually you know

but um you know AMD actually you know has of of course been in in CPUs and I

has of of course been in in CPUs and I kind of mentioned like there's that's

kind of mentioned like there's that's what's in your device you know your

what's in your device you know your notebook or your desktop versus what's

notebook or your desktop versus what's in the cloud so we do server and client

in the cloud so we do server and client CPUs we also have I mentioned um you

CPUs we also have I mentioned um you know the gpus that are very focused on

know the gpus that are very focused on really high performant uh Ai and that's

really high performant uh Ai and that's the Mi Instinct line and then we have

the Mi Instinct line and then we have the radio line which is still doing

the radio line which is still doing gaming which by the way of course in

gaming which by the way of course in gaming there is going to be some AI

gaming there is going to be some AI there but we also have um even more

there but we also have um even more special I won't go into the primer on

special I won't go into the primer on that but uh products from uh the zyink

that but uh products from uh the zyink acquisition uh which is a product called

acquisition uh which is a product called fpgas and adaptive so so's and and these

fpgas and adaptive so so's and and these are products that are in many many

are products that are in many many embedded products and the reason why I

embedded products and the reason why I want to first you know share the the

want to first you know share the the breath of the product portfolio is

breath of the product portfolio is because also it it's uh it's in many

because also it it's uh it's in many different products that actually you may

different products that actually you may not think you're touching it but you're

not think you're touching it but you're probably using an AMD product every day

probably using an AMD product every day um because we're in cars um we're in

um because we're in cars um we're in infrastructure we're in Communications

infrastructure we're in Communications infrastructure both wir and wireless and

infrastructure both wir and wireless and then of course we're in in like the more

then of course we're in in like the more traditional Computing things we are in

traditional Computing things we are in healthcare um so so it's quite Broad and

healthcare um so so it's quite Broad and the reason why I also mentioned that is

the reason why I also mentioned that is because you know AI is going to um I

because you know AI is going to um I think you know just you know

think you know just you know revolutionize and really en enhance um

revolutionize and really en enhance um everybody's everyday lives and of course

everybody's everyday lives and of course we have to be mindful about the negative

we have to be mindful about the negative things but um you know we do play across

things but um you know we do play across all of that and there's different

all of that and there's different software Dynamics around that um and I

software Dynamics around that um and I think that you know we think of this as

think that you know we think of this as um important on multiple levels because

um important on multiple levels because it really does you know hearing about

it really does you know hearing about the healthcare that's a perfect example

the healthcare that's a perfect example of you know you want to enable that

of you know you want to enable that Innovation while you're doing the proper

Innovation while you're doing the proper things on privacy and all the others and

things on privacy and all the others and I believe in this comment of the genius

I believe in this comment of the genius of the end it's like you can do the

of the end it's like you can do the right thing and still be a very

right thing and still be a very successful profitable business creating

successful profitable business creating value for society as a whole and

value for society as a whole and competing very rigorously right and I

competing very rigorously right and I think that's the goal of government

think that's the goal of government ensuring that sort of happening and when

ensuring that sort of happening and when it doesn't and Comm measure to sort of

it doesn't and Comm measure to sort of the risk factors maybe you know playing

the risk factors maybe you know playing a stronger role but what you really want

a stronger role but what you really want to see is that competition that you know

to see is that competition that you know you know and again the The Innovation

you know and again the The Innovation because that drives cost that drives

because that drives cost that drives greater value and so forth and um and

greater value and so forth and um and you know again we take an open approach

you know again we take an open approach that we try to enable an ecosystem as

that we try to enable an ecosystem as opposed to a more proprietary you know

opposed to a more proprietary you know locked in approach so excellent all

locked in approach so excellent all right so let's try and put this in a

right so let's try and put this in a little bit of perspective overall um

little bit of perspective overall um those of us who've been around for a

those of us who've been around for a while we have seen Tech come and go

while we have seen Tech come and go right we used to have mainframes and

right we used to have mainframes and then there were the rise of personal

then there were the rise of personal computers and now we have um we have

computers and now we have um we have more on our phones than um whole rooms

more on our phones than um whole rooms of mainframes used to be that we carry

of mainframes used to be that we carry around with us every day um so is

around with us every day um so is this um AI Hardware right now is this

this um AI Hardware right now is this cycle really any different um is it

cycle really any different um is it really any different than any of the

really any different than any of the other ones that have happened in the

other ones that have happened in the last 20 or 40 years and how and why

last 20 or 40 years and how and why might it be different so um to maer let

might it be different so um to maer let me start with you and let me say feel

me start with you and let me say feel free to all my panelists to to jump in

free to all my panelists to to jump in but muzzer kick us off

but muzzer kick us off um so I think the short answer is yes

um so I think the short answer is yes from my perspective from what I've uh

from my perspective from what I've uh observed um I like to think about or

observed um I like to think about or kind of describe this as um if if you

kind of describe this as um if if you guys have heard of the Canan effect

guys have heard of the Canan effect Canan effect is this idea that inflation

Canan effect is this idea that inflation is experienced by different parts of

is experienced by different parts of society differently so uh people closest

society differently so uh people closest to the money printer the central banks

to the money printer the central banks the primary dealers the insurance

the primary dealers the insurance companies Etc they get the value of the

companies Etc they get the value of the marginal dollar uh before anyone else

marginal dollar uh before anyone else and therefore they they're sort of this

and therefore they they're sort of this like wealth transfer from from society

like wealth transfer from from society to you know to to the financial sector

to you know to to the financial sector and I think that um and that's sort of

and I think that um and that's sort of an exponential function and and I think

an exponential function and and I think something like that is happening in the

something like that is happening in the AI space especially with uh companies

AI space especially with uh companies that have strong AI capabilities either

that have strong AI capabilities either the uh the the best model or the best

the uh the the best model or the best data lake or um or uh creating the best

data lake or um or uh creating the best AI Hardware is that they have this sort

AI Hardware is that they have this sort of feedback loop that they can they can

of feedback loop that they can they can sort of generate the next model faster

sort of generate the next model faster the next the next powerful model faster

the next the next powerful model faster better at accelerating rate uh versus

better at accelerating rate uh versus their competition so there is that

their competition so there is that feedback loop that I think is is

feedback loop that I think is is different and you can observe this by a

different and you can observe this by a lot of startups having to do really

lot of startups having to do really rapid pivots uh since the Advent of gpt3

rapid pivots uh since the Advent of gpt3 in November 2022 uh and then every

in November 2022 uh and then every subsequent release um people are

subsequent release um people are realizing that they sort of have to rip

realizing that they sort of have to rip out that intelligent portion that sort

out that intelligent portion that sort of was their core value and just replace

of was their core value and just replace it with a strong AA model and so I think

it with a strong AA model and so I think that is different where larger players

that is different where larger players not only have distribution but they have

not only have distribution but they have the speed that was usually the

the speed that was usually the characteristic of a small company got it

characteristic of a small company got it I could say I I also agree it's

I could say I I also agree it's different and I'll I'll point it three

different and I'll I'll point it three um three aspects you know one is you

um three aspects you know one is you know and showing my age right I started

know and showing my age right I started at a school in 82 so I've seen a lot of

at a school in 82 so I've seen a lot of stuff I've seen you know the the PC

stuff I've seen you know the the PC generation the mobile you know social

generation the mobile you know social media everything um you know I I think

media everything um you know I I think you know Percy and Andrew talked about

you know Percy and Andrew talked about this this is a very general purpose

this this is a very general purpose technology so the fact that it's so

technology so the fact that it's so Broad and it can appli for everything

Broad and it can appli for everything from healthare to forecasting to you

from healthare to forecasting to you know uh entertainment to

know uh entertainment to creativity infrastructure smart cities

creativity infrastructure smart cities um autonomous driving it's just so broad

um autonomous driving it's just so broad so so that's one factor that's different

so so that's one factor that's different the other factor is different we have

the other factor is different we have some of the other disruptions right so

some of the other disruptions right so if you had you know 56 years of MW and

if you had you know 56 years of MW and the power of semiconductors is

the power of semiconductors is incredible so algorithms and things that

incredible so algorithms and things that weren't useful before but now with the

weren't useful before but now with the computing power that you can have like

computing power that you can have like you said on an on on your phone or on a

you said on an on on your phone or on a device that you carry is is phenomenal

device that you carry is is phenomenal and we have the internet so the speed at

and we have the internet so the speed at which you know information travels the

which you know information travels the I've never seen a concept in you know

I've never seen a concept in you know Academia get commercialized so fast in

Academia get commercialized so fast in in the 42 years of my career right and

in the 42 years of my career right and the last thing is actually also just

the last thing is actually also just adoption because I remember it took a

adoption because I remember it took a long time for the internet to really I I

long time for the internet to really I I remember hearing about mosaic

remember hearing about mosaic you know but you have to be an Insider

you know but you have to be an Insider geek kind of person to know that and

geek kind of person to know that and then slowly there was AOL and all stuff

then slowly there was AOL and all stuff I mean but this is like and and it's

I mean but this is like and and it's true there was really something before

true there was really something before chap gbt but that is an amazing

chap gbt but that is an amazing phenomena right where my kids

phenomena right where my kids immediately within months were asking me

immediately within months were asking me questions about that they never ask me

questions about that they never ask me questions about my work that in itself

questions about my work that in itself is like you know unprecedented right um

is like you know unprecedented right um but you know not just in the US around

but you know not just in the US around the world right and there are people

the world right and there are people experimenting with it so so that human

experimenting with it so so that human um awareness and adoption and and that

um awareness and adoption and and that and the conversation that's amazing I

and the conversation that's amazing I have not seen that you know usually it

have not seen that you know usually it takes years before a new technology hits

takes years before a new technology hits people's um Consciousness like like it

people's um Consciousness like like it has so yep those are my three reasons

has so yep those are my three reasons why I think this time it's different yep

why I think this time it's different yep famous last words sometimes but Chris

famous last words sometimes but Chris did you have thought yeah I I think the

did you have thought yeah I I think the other part of this too I guess we're all

other part of this too I guess we're all in like Universal agreement here I wish

in like Universal agreement here I wish I had more success though in

I had more success though in communicating to my family what I do

communicating to my family what I do like if you ask my mom what I do she

like if you ask my mom what I do she like he works on

like he works on computers and then it would be like you

computers and then it would be like you had to go all the way to Brazil to fix

had to go all the way to Brazil to fix someone's computer and I'm like yeah Mom

someone's computer and I'm like yeah Mom I wish I just told them to reboot you

I wish I just told them to reboot you know but you know if you put that aside

know but you know if you put that aside there's a huge difference now with like

there's a huge difference now with like AI Hardware compared to something like a

AI Hardware compared to something like a mobile phone right where you'd have a

mobile phone right where you'd have a phone you'd have an OS and you'd have a

phone you'd have an OS and you'd have a fairly vertical stack with with there's

fairly vertical stack with with there's a ton of autonomy and flexibility with

a ton of autonomy and flexibility with AI Hardware today where you have

AI Hardware today where you have projects such as Ray to really help

projects such as Ray to really help orchestrate AI workloads you have uh AI

orchestrate AI workloads you have uh AI Frameworks like pytorch that can really

Frameworks like pytorch that can really help to create that neutrality you have

help to create that neutrality you have virtualization software like we provide

virtualization software like we provide that provides a lot of flexibility so

that provides a lot of flexibility so you there's tons of ways and lots of

you there's tons of ways and lots of different approaches you can take to

different approaches you can take to decouple applications from Hardware so

decouple applications from Hardware so it's a much different phenomenon than

it's a much different phenomenon than what you saw with a mobile phone in the

what you saw with a mobile phone in the past so I I'll just say I have no

past so I I'll just say I have no overarching answer to is AI different

overarching answer to is AI different than any past technology I'm happy to

than any past technology I'm happy to take a weit and see approach but I do

take a weit and see approach but I do think we are seeing something that is

think we are seeing something that is different from the last 20 years

different from the last 20 years particularly with re with respect to

particularly with re with respect to chips and Hardware which is that uh if

chips and Hardware which is that uh if you're not a practitioner this may sound

you're not a practitioner this may sound uh you know strange and if you are I

uh you know strange and if you are I guess it'll be boring but the last 20

guess it'll be boring but the last 20 years were really defined by a use of

years were really defined by a use of general purpose commodity Hardware in

general purpose commodity Hardware in the data center it was not about

the data center it was not about specialized chips for each workload if

specialized chips for each workload if you whether you were running a social

you whether you were running a social media application or an Enterprise

media application or an Enterprise application or as a software as a

application or as a software as a service business probably you were

service business probably you were running on roughly similar hardware and

running on roughly similar hardware and AI is really taking a different approach

AI is really taking a different approach in that it is fundamentally accelerated

in that it is fundamentally accelerated by gpus and even more specialized

by gpus and even more specialized hardware and it is really I think

hardware and it is really I think driving a Resurgence of uh what we might

driving a Resurgence of uh what we might call like an accelerator right dedicated

call like an accelerator right dedicated hardware for a specific task and I just

hardware for a specific task and I just think from the you know perspective of

think from the you know perspective of chips like that is something that's just

chips like that is something that's just different from how we you know the

different from how we you know the median business even the median Tech

median business even the median Tech business was operating five or 10 years

business was operating five or 10 years ago absolutely let me um let me go on to

ago absolutely let me um let me go on to my next question um what are the market

my next question um what are the market dynamics we're thinking about these

dynamics we're thinking about these chips that are now we're hearing more

chips that are now we're hearing more and more important that this is

and more important that this is different um you may need vast

different um you may need vast quantities of them um and we're here to

quantities of them um and we're here to talk about competition in markets so

talk about competition in markets so what are the market dynamics when we're

what are the market dynamics when we're thinking about these chips how do chips

thinking about these chips how do chips affect access to Capital how does access

affect access to Capital how does access to Capital affect access to chips um

to Capital affect access to chips um blanch can you start us off yes thank

blanch can you start us off yes thank you very much yes I will begin with

you very much yes I will begin with Basics so you've understand I'm not the

Basics so you've understand I'm not the the technical person in the panel but I

the technical person in the panel but I will give you my perspective on on how

will give you my perspective on on how we see the market of chips from really a

we see the market of chips from really a company that needs them uh so beginning

company that needs them uh so beginning with the basics uh the requirements for

with the basics uh the requirements for training are not the same as for

training are not the same as for inference it has been said in more

inference it has been said in more technical words but that's important to

technical words but that's important to recall the basics and uh servicing the

recall the basics and uh servicing the chips that are the ones that work for

chips that are the ones that work for training uh requires um some criteria

training uh requires um some criteria which which is um the most powerful

which which is um the most powerful chips so we speak about flops uh a chips

chips so we speak about flops uh a chips with lots of memory and also so those

with lots of memory and also so those chips today uh the h00 the H 200 that

chips today uh the h00 the H 200 that will come they are pretty rare that's

will come they are pretty rare that's the first point and also the third

the first point and also the third aspect is um for some reason that I

aspect is um for some reason that I cannot explain but this is what we

cannot explain but this is what we observe for some reason having those

observe for some reason having those chips is not enough you need a cloud

chips is not enough you need a cloud provider that has a very good

provider that has a very good infrastructure uh to assemble them by

infrastructure uh to assemble them by eight and to uh and to have uh to to

eight and to uh and to have uh to to assemble them in a way that uh does that

assemble them in a way that uh does that that creates a good connectivity between

that creates a good connectivity between the chips otherwise the training is

the chips otherwise the training is super long and it just fails or it costs

super long and it just fails or it costs too much and and that is something that

too much and and that is something that so the chips are rare and the providers

so the chips are rare and the providers that can uh the the the companies that

that can uh the the the companies that can assemble them and make them work in

can assemble them and make them work in a way that is efficient for training are

a way that is efficient for training are even more rare so um so if you have

even more rare so um so if you have providers that do not combine all those

providers that do not combine all those criteria you can do some inference with

criteria you can do some inference with them you can do some other things

them you can do some other things someone spoke about Rag and that's true

someone spoke about Rag and that's true you can do a lot of things with with

you can do a lot of things with with less power with less memory and and also

less power with less memory and and also maybe with some infrastructure that is

maybe with some infrastructure that is less efficient but you cannot do

less efficient but you cannot do training the other thing that is

training the other thing that is important for training is that you need

important for training is that you need a huge number of chips and you need them

a huge number of chips and you need them at the same place so that means that if

at the same place so that means that if you don't have them at the same place

you don't have them at the same place you will be able to train several models

you will be able to train several models of different sizes adapted maybe to the

of different sizes adapted maybe to the chips that you have available in the

chips that you have available in the same data center but you won't be able

same data center but you won't be able to build a huge model uh with your chips

to build a huge model uh with your chips scattered in different places so what

scattered in different places so what that means really practically is that

that means really practically is that you need a very big ticket of cash uh to

you need a very big ticket of cash uh to the same provider to be able to train

the same provider to be able to train big models and so uh if you don't have

big models and so uh if you don't have catch you don't have that but if you

catch you don't have that but if you don't have that you also have problems

don't have that you also have problems convincing your investors to get your

convincing your investors to get your next fundraising because you have

next fundraising because you have problems convincing them that you are a

problems convincing them that you are a realistic actor for for the future so

realistic actor for for the future so that's kind of what we're struggling

that's kind of what we're struggling with uh struggling not not so much

with uh struggling not not so much because we've managed so far uh and also

because we've managed so far uh and also because um maybe we took the opportunity

because um maybe we took the opportunity to not be being able to train all the

to not be being able to train all the time the biggest models uh to make sure

time the biggest models uh to make sure we trained the best smaller models which

we trained the best smaller models which also has some virtues because not all

also has some virtues because not all use cases

use cases requires the biggest models and it's

requires the biggest models and it's better for environment it's also been

better for environment it's also been said already but that's maybe how how

said already but that's maybe how how the lscape is um as of today so as of

the lscape is um as of today so as of today what we are experiencing and that

today what we are experiencing and that will be my conclusion at this point is a

will be my conclusion at this point is a massive bottleneck both on the provision

massive bottleneck both on the provision of chips and on the provision by the the

of chips and on the provision by the the the few companies that know how to

the few companies that know how to manage them in the right way to be able

manage them in the right way to be able to train models H and that means that

to train models H and that means that there is a cue and there is a cue for a

there is a cue and there is a cue for a company like ours on training but there

company like ours on training but there is also a cue for customers of ours that

is also a cue for customers of ours that would like to have their own

would like to have their own infrastructure for inference this time

infrastructure for inference this time so that reduces the possibility of

so that reduces the possibility of distribution on the side of of customers

distribution on the side of of customers uh to do on premises for example and

uh to do on premises for example and that also probably reduces the

that also probably reduces the possibility of new companies to to start

possibility of new companies to to start from scratch and convince all the all

from scratch and convince all the all these actors to provide them fast with a

these actors to provide them fast with a lot of

lot of computes all right well a a bottleneck

computes all right well a a bottleneck of chips you need a only one major

of chips you need a only one major provider sounds like an antitrust

provider sounds like an antitrust problem to me of course I'm an antitrust

problem to me of course I'm an antitrust lawyer so I see that I I feel compelled

lawyer so I see that I I feel compelled so say Victor we're we're coming to your

so say Victor we're we're coming to your rescue right we're going to we're going

rescue right we're going to we're going to like reduce that scarcity reduce the

to like reduce that scarcity reduce the TCO give you Choice um and you know look

TCO give you Choice um and you know look let's face it you know the Mi 300X which

let's face it you know the Mi 300X which is getting stood up in like it's public

is getting stood up in like it's public you know Azure and a lot of the public

you know Azure and a lot of the public you know infrastructure is going to get

you know infrastructure is going to get out there we're getting there um and we

out there we're getting there um and we really do you know have the leadership

really do you know have the leadership in inference exactly that for large

in inference exactly that for large language models you know we bench talk

language models you know we bench talk on things like llama and you mentioned

on things like llama and you mentioned also the distribution that's why we

also the distribution that's why we saying network equipment we also do have

saying network equipment we also do have networking equipment but we're also

networking equipment but we're also partnering with broadcom and with others

partnering with broadcom and with others actually to enable an open uh

actually to enable an open uh interconnect because you know it is true

interconnect because you know it is true you need powerful you know compute nodes

you need powerful you know compute nodes but you also for these large models you

but you also for these large models you need to have the connectivity because

need to have the connectivity because you could lose a lot of porns through

you could lose a lot of porns through that um you know now at the same time

that um you know now at the same time there are more Modz models and you know

there are more Modz models and you know one thing I didn't talk about earlier I

one thing I didn't talk about earlier I talked about how AMD spans end to end um

talked about how AMD spans end to end um we're in aipc so in your laptop we have

we're in aipc so in your laptop we have acceleration within that right so

acceleration within that right so there's no you don't have the same um

there's no you don't have the same um connectivity issues and that's going to

connectivity issues and that's going to go in in to to a large extent help

go in in to to a large extent help democratize maybe if you will like you

democratize maybe if you will like you know get get broader access but it

know get get broader access but it wouldn't be so for the really extremely

wouldn't be so for the really extremely large models that they they need the

large models that they they need the large infrastructure right but I I

large infrastructure right but I I really think this is goes to the heart

really think this is goes to the heart of you know if they're innovators like

of you know if they're innovators like mral is Bal neck what they need is they

mral is Bal neck what they need is they need a really active Lively Market at

need a really active Lively Market at that net foundational level and that's

that net foundational level and that's what we truly mean to provide and that

what we truly mean to provide and that is where if Beyond you know we're we're

is where if Beyond you know we're we're going to compete vigorously we have

going to compete vigorously we have leadership and inference right um but if

leadership and inference right um but if there are things that are not equal in

there are things that are not equal in competition or things like that's the

competition or things like that's the thing where you have to be careful of

thing where you have to be careful of and then I think you know outside of

and then I think you know outside of Regulation there's other things that I

Regulation there's other things that I think could be done by the government um

think could be done by the government um to helps for Innovation but anyway I

to helps for Innovation but anyway I just I just absolutely wanted to say

just I just absolutely wanted to say help is on the way I hope you look at

help is on the way I hope you look at AMD so all right well let me jump to um

AMD so all right well let me jump to um talking about the role of government

talking about the role of government here we potentially have a bottleneck we

here we potentially have a bottleneck we potentially have some super large

potentially have some super large players having super large roles um in

players having super large roles um in this industry um government can do sort

this industry um government can do sort of two things uh broadly one they can

of two things uh broadly one they can put money on the table they can put

put money on the table they can put subsidies on the table they can

subsidies on the table they can contribute in that way the other way is

contribute in that way the other way is through regulation um Alex um from the

through regulation um Alex um from the FTC perspective you said a minute ago

FTC perspective you said a minute ago you're going to have a wait and seee

you're going to have a wait and seee approach but I don't think that was a

approach but I don't think that was a global statement um correct tell us tell

global statement um correct tell us tell us a little bit about um what the ft is

us a little bit about um what the ft is thinking about the role of government in

thinking about the role of government in what is potentially a bottlenecked

what is potentially a bottlenecked environment controlled by a couple large

environment controlled by a couple large players yeah so the ftc's role is as an

players yeah so the ftc's role is as an enforcer of the nation's competition and

enforcer of the nation's competition and consumer protection laws and the first

consumer protection laws and the first order belief we have is that there is no

order belief we have is that there is no AI exemption to the rules that are on

AI exemption to the rules that are on the book existing competition laws apply

the book existing competition laws apply to this sector like they would any other

to this sector like they would any other I think the uh a caner said it well this

I think the uh a caner said it well this morning if you use an algorithm to

morning if you use an algorithm to collusively price fix that's is illegal

collusively price fix that's is illegal if you didn't use an algorithm right if

if you didn't use an algorithm right if you attempt to monopolize you know a

you attempt to monopolize you know a market whether it's for AI models or in

market whether it's for AI models or in the AI supply chain that is already a

the AI supply chain that is already a violation of our competition laws so I

violation of our competition laws so I think that that's the first order answer

think that that's the first order answer is we look to vigorously enforce the

is we look to vigorously enforce the laws that are on the books um in January

laws that are on the books um in January of this year we announced a 6B study uh

of this year we announced a 6B study uh which is an authority we have for a

which is an authority we have for a non-law enforcement investigation uh to

non-law enforcement investigation uh to better understand a market looking at

better understand a market looking at Investments and Partnerships between uh

Investments and Partnerships between uh large model developers and large comb

large model developers and large comb and Cloud providers and the focus of

and Cloud providers and the focus of that is to understand what is the nature

that is to understand what is the nature of these Investments and Partnerships

of these Investments and Partnerships and how do they impact the competitive

and how do they impact the competitive Dynamics in those marketplaces so that

Dynamics in those marketplaces so that that's something we're actively doing

that's something we're actively doing and I I don't have any results to share

and I I don't have any results to share but you know we're actively looking to

but you know we're actively looking to understand how the market is shaped by

understand how the market is shaped by uh those types of

uh those types of structures great thank you I'm going to

structures great thank you I'm going to flip back to blunch real briefly because

flip back to blunch real briefly because um Europe doesn't always do things

um Europe doesn't always do things exactly the way we do them in the US um

exactly the way we do them in the US um maybe you can just give a little

maybe you can just give a little International perspective in terms of

International perspective in terms of the role of government um in in France

the role of government um in in France or in

or in Europe I think it's it's the reputation

Europe I think it's it's the reputation indeed of Europe to be very good at

indeed of Europe to be very good at regulation I don't know if I I think

regulation I don't know if I I think it's great but it is what it is but for

it's great but it is what it is but for some purposes it is indeed great so I

some purposes it is indeed great so I think there are indeed two aspects to

think there are indeed two aspects to the topic one of them is really

the topic one of them is really encouraging this Innovation putting

encouraging this Innovation putting money on the table creating the

money on the table creating the universities and and the schools that

universities and and the schools that will accompany population to to ramp up

will accompany population to to ramp up to this Innovation make sure that there

to this Innovation make sure that there is no one left behind into this adoption

is no one left behind into this adoption and and also that the persons where

and and also that the persons where which have the jobs that are

which have the jobs that are revolutionized by this technical

revolutionized by this technical Revolution are accompanied to to to

Revolution are accompanied to to to change their way to work so I think that

change their way to work so I think that there's a lot to do for governments on

there's a lot to do for governments on the positive side uh and I think that

the positive side uh and I think that France is doing a lot on this and and

France is doing a lot on this and and this is maybe one of the very good

this is maybe one of the very good reasons why M TR is here today because a

reasons why M TR is here today because a decade ago the French government has

decade ago the French government has started to put billions on the table and

started to put billions on the table and and to make sure that there were some

and to make sure that there were some centers in France and Paris where

centers in France and Paris where students would be very good at Ai and

students would be very good at Ai and there would be some uh also some

there would be some uh also some welcoming some uh some of these

welcoming some uh some of these hyperscalers in Paris to have their Hub

hyperscalers in Paris to have their Hub their AI Hub in Paris to to keep a

their AI Hub in Paris to to keep a French ecosystem Lively also thanks to

French ecosystem Lively also thanks to us companies to to be frank uh so that

us companies to to be frank uh so that what happened and and a week ago our

what happened and and a week ago our French president maon announced some

French president maon announced some other hundreds of millions put on the

other hundreds of millions put on the table uh for um funding AI Innovation

table uh for um funding AI Innovation including chips uh and for training also

including chips uh and for training also um more people on AI so I think this is

um more people on AI so I think this is a very good thing to do and and it's

a very good thing to do and and it's very welcome in every country for sure

very welcome in every country for sure because really the purpose is to to

because really the purpose is to to leave no one behind uh and I'm sure

leave no one behind uh and I'm sure there is a lot of money in the US to be

there is a lot of money in the US to be put on that too on the other side of the

put on that too on the other side of the of the topic which is more the

of the topic which is more the regulation in the mere sense of the term

regulation in the mere sense of the term uh there is this AI act in Europe some

uh there is this AI act in Europe some might think it's going too soon because

might think it's going too soon because because we don't know yet what we

because we don't know yet what we regulating but it has the merits of uh

regulating but it has the merits of uh putting some uh some problematics on the

putting some uh some problematics on the table so that's always interesting but

table so that's always interesting but um so yeah there is a there is a thing

um so yeah there is a there is a thing which is it's going so fast that

which is it's going so fast that Regulators want to go fast which is

Regulators want to go fast which is really

really admirable but it's also a risk that if

admirable but it's also a risk that if they go too fast they regulate too much

they go too fast they regulate too much not on the right thing so maybe this is

not on the right thing so maybe this is a risk there and on more on the on the

a risk there and on more on the on the competition side all the European

competition side all the European authorities have been very active in

authorities have been very active in trying to ramp up on the topic and

trying to ramp up on the topic and understand and that's great because

understand and that's great because someday they will need to react and they

someday they will need to react and they will need to have all this science uh in

will need to have all this science uh in in in this area and honestly just by

in in this area and honestly just by answering them I've learned so much into

answering them I've learned so much into the that I think it's a very good

the that I think it's a very good initiative there are also been some

initiative there are also been some stronger initiatives into looking into

stronger initiatives into looking into Partnerships and here M believes it's a

Partnerships and here M believes it's a risk because pure players like us we

risk because pure players like us we need Partnerships to exist we need we

need Partnerships to exist we need we are not integrated on the value stack so

are not integrated on the value stack so we need to partner with hyperscalers for

we need to partner with hyperscalers for compute for distribution for almost

compute for distribution for almost everything and if we cannot do that uh

everything and if we cannot do that uh there is a risk that we just do not

there is a risk that we just do not exist so here we tend to think that

exist so here we tend to think that there is a risk risk that by reacting

there is a risk risk that by reacting too fast it could also create some side

too fast it could also create some side effects that could have contrary effects

effects that could have contrary effects to what competition law is here for so

to what competition law is here for so that's kind of the dynamic knowing the

that's kind of the dynamic knowing the topic being ready to react and to and to

topic being ready to react and to and to and regulate and and trying to

and regulate and and trying to anticipate the side effects which is

anticipate the side effects which is very difficult given how fast it's

very difficult given how fast it's moving got it thank you anyone else want

moving got it thank you anyone else want to jump in on the role of government

to jump in on the role of government Chris I would just say you know a couple

Chris I would just say you know a couple things that you know we're we're we're

things that you know we're we're we're very supportive of what's already

very supportive of what's already happened with the president's recent

happened with the president's recent executive order um as well as even the

executive order um as well as even the NY framework which is something our

NY framework which is something our internal AI policy also points to these

internal AI policy also points to these are very effective guidelines for

are very effective guidelines for steering the safe and secure adoption of

steering the safe and secure adoption of AI and it focuses on areas such as being

AI and it focuses on areas such as being able to uh AB test different AI models

able to uh AB test different AI models being able to have choice and this is

being able to have choice and this is what's Driven our both internal and

what's Driven our both internal and external approach to having a AI

external approach to having a AI platform the other thing I would just

platform the other thing I would just mention is like I've seen this a lot

mention is like I've seen this a lot like my career the last decade I've key

like my career the last decade I've key part of my role has been bringing new

part of my role has been bringing new technologies to Market and where I see a

technologies to Market and where I see a little bit of a gap in terms of support

little bit of a gap in terms of support from the government is the government's

from the government is the government's been very good in terms of trying to

been very good in terms of trying to promote Innovation but hasn't invested

promote Innovation but hasn't invested equally in promoting adoption of

equally in promoting adoption of innovation so sometimes getting that

innovation so sometimes getting that first mover agency to to to adopt

first mover agency to to to adopt technology is where that bottleneck

technology is where that bottleneck exists because if I'm a CIO of an agency

exists because if I'm a CIO of an agency I want to see which of my peers has

I want to see which of my peers has adopted first so incentivizing adoption

adopted first so incentivizing adoption is just as important as incentivizing

is just as important as incentivizing Innovation interesting interesting

Innovation interesting interesting thanks I wouldn't have guessed that like

thanks I wouldn't have guessed that like as Victor was saying everyone used chat

as Victor was saying everyone used chat GPT the week after it came out and um

GPT the week after it came out and um but maybe but there's a difference

but maybe but there's a difference between all of us typing something in

between all of us typing something in randomly um and really using it in a

randomly um and really using it in a business context where they're going to

business context where they're going to be looking at what are their competitors

be looking at what are their competitors doing a big part of my job last year was

doing a big part of my job last year was saying no to product teams wanting to

saying no to product teams wanting to embed chat GPT Integrations into

embed chat GPT Integrations into products got it okay um let's talk a

products got it okay um let's talk a little bit about vertical integration

little bit about vertical integration we've talked a little bit about the

we've talked a little bit about the stack chips on the bottom chat GPT on

stack chips on the bottom chat GPT on the top um and um and if the chip

the top um and um and if the chip companies um one of which is a very

companies um one of which is a very dominant position um in some of these

dominant position um in some of these areas um if it keeps integrating up the

areas um if it keeps integrating up the stack you can have a completely

stack you can have a completely integrated product which might be great

integrated product which might be great in some ways and might be horrible in

in some ways and might be horrible in some others um so um it might be good

some others um so um it might be good for uh it might be horrible or good at

for uh it might be horrible or good at different levels customers might have a

different levels customers might have a view um chip makers might have a view

view um chip makers might have a view people who are on the middle of the

people who are on the middle of the stack might have um a different view so

stack might have um a different view so um vertical integration what should

um vertical integration what should customers be thinking about what should

customers be thinking about what should government Regulators be thinking about

government Regulators be thinking about what should other companies be thinking

what should other companies be thinking about um Victor want to start sure and

about um Victor want to start sure and and again I'm going to pick up on um how

and again I'm going to pick up on um how it was already described several times

it was already described several times about you know it's you need to connect

about you know it's you need to connect up all these really powerful servers you

up all these really powerful servers you know GPU rich and also there CPUs in

know GPU rich and also there CPUs in there um you know now if you want to

there um you know now if you want to have choice and you want to have reduce

have choice and you want to have reduce the scarcity problem and therefore

the scarcity problem and therefore lowering costs and improving access to

lowering costs and improving access to innovators and so forth they need to

innovators and so forth they need to interoperate right so if the

interoperate right so if the connectivity in the network is

connectivity in the network is proprietary and closed and it's really

proprietary and closed and it's really hard for AMD operate with competition by

hard for AMD operate with competition by the way and AMD inter operates with you

the way and AMD inter operates with you know like traditionally with um you know

know like traditionally with um you know our comp competition and CPUs and and

our comp competition and CPUs and and others then that's challenge right

others then that's challenge right because that's that's blocking um a need

because that's that's blocking um a need from consumers and blocking ultimately

from consumers and blocking ultimately then you know progress right in

then you know progress right in Innovation and and Global um you know

Innovation and and Global um you know creating new value reducing the marginal

creating new value reducing the marginal cost of value and so forth right so I

cost of value and so forth right so I think there there's a you know because

think there there's a you know because there's a lot of discussion around

there's a lot of discussion around software and absolutely you have to look

software and absolutely you have to look at the software um interfaces and and

at the software um interfaces and and the layers in the stack but you also

the layers in the stack but you also have to think about the hardware and

have to think about the hardware and again I I I'm back to the there's

again I I I'm back to the there's nothing inherently wrong about vertical

nothing inherently wrong about vertical integration that happens a lot but if if

integration that happens a lot but if if it's closed and vertically integrated

it's closed and vertically integrated and then there's a extremely you know

and then there's a extremely you know persistent um position there then then

persistent um position there then then that that's a barrier right for for for

that that's a barrier right for for for customer choice for Innovation for

customer choice for Innovation for driving down cost for you know enabling

driving down cost for you know enabling you new value creation models so right

you new value creation models so right anyone else want to jump in leser I'll

anyone else want to jump in leser I'll uh so the way I think about this is that

uh so the way I think about this is that there's kind of a problem at the high

there's kind of a problem at the high end where a sort of dominant AI

end where a sort of dominant AI capability whether it's Hardware models

capability whether it's Hardware models or data could be used to reduce choice

or data could be used to reduce choice in somewhere else so one specific

in somewhere else so one specific example is um if you wanted to start a

example is um if you wanted to start a healthc care company that uses the

healthc care company that uses the highest end

highest end Ai and it has to be hit by compliant

Ai and it has to be hit by compliant there's only one cloud provider that can

there's only one cloud provider that can do that for you

do that for you uh if you want access to gpus well 40%

uh if you want access to gpus well 40% of nvidia's Revenue came from the

of nvidia's Revenue came from the hyperscalers you got to go to a large

hyperscalers you got to go to a large Club provider usually you won't have

Club provider usually you won't have access to those gpus if Google decides

access to those gpus if Google decides to say that hey if you want access to

to say that hey if you want access to the pedabytes and pedabytes of data that

the pedabytes and pedabytes of data that is generated every day by YouTube if you

is generated every day by YouTube if you want to if you want to train on that you

want to if you want to train on that you have to use

have to use gcp you have one choice so I think I

gcp you have one choice so I think I think um I think the vertical

think um I think the vertical integration I think one way to think

integration I think one way to think about is it's good in a way that gives

about is it's good in a way that gives you kind of a a solution that just works

you kind of a a solution that just works um but I think the the real problem is

um but I think the the real problem is where if you wanted to use a strong AI

where if you wanted to use a strong AI capability and that reduces choice in a

capability and that reduces choice in a different layer that's really where it's

different layer that's really where it's it's a little bit Pro problematic and

it's a little bit Pro problematic and that maybe that's a way that you know

that maybe that's a way that you know for example we can make proprietary

for example we can make proprietary models even widely distributed right or

models even widely distributed right or um maybe make uh compute or the

um maybe make uh compute or the distribution of compute somehow more

distribution of compute somehow more distributed so the role of government

distributed so the role of government then these are being these locked in

then these are being these locked in models what do we think I mean my

models what do we think I mean my perspective just as an engineer is

perspective just as an engineer is modularity tends to be good for

modularity tends to be good for Innovation because when you have a

Innovation because when you have a modular system you can have competition

modular system you can have competition for each component and more competitors

for each component and more competitors you know whether it's the same company

you know whether it's the same company competing at multiple layers or entirely

competing at multiple layers or entirely different companies competing at

different companies competing at different layers gets you more

different layers gets you more Innovation as they each strive to

Innovation as they each strive to perform you know the role of their you

perform you know the role of their you know component well and interoperate

know component well and interoperate with the other layers you know I think

with the other layers you know I think that's a tends to be a pro-innovation

that's a tends to be a pro-innovation Pro competition recipe when you have

Pro competition recipe when you have that

that modularity got it okay without saying

modularity got it okay without saying whether the government should do it it's

whether the government should do it it's still from an engineering standpoint yes

still from an engineering standpoint yes just yeah just as an engineer

just yeah just as an engineer that's all right I'm getting a sign that

that's all right I'm getting a sign that we have one minute left I'm going to go

we have one minute left I'm going to go to my last question we're going to call

to my last question we're going to call this a lightning round so everybody gets

this a lightning round so everybody gets you know two

you know two sentences 10 years from now crystal ball

sentences 10 years from now crystal ball what is the uh the AI Hardware uh market

what is the uh the AI Hardware uh market look like dominated by a one or two

look like dominated by a one or two players uh has it integrated all the way

players uh has it integrated all the way up the stack um has government

up the stack um has government intervened has the world fractured and

intervened has the world fractured and they've done something different in EU

they've done something different in EU as opposed to somewhere else um Chris

as opposed to somewhere else um Chris yeah I think it's going to be Lively and

yeah I think it's going to be Lively and robust there's going to be tons of

robust there's going to be tons of players you already see more accelerator

players you already see more accelerator vendors today than than we've

vendors today than than we've historically had with x86 CPU vendors

historically had with x86 CPU vendors and the thing that's going to really

and the thing that's going to really pick this up is going to be the

pick this up is going to be the prevalence of smaller domain specific AI

prevalence of smaller domain specific AI models that's what's going to really be

models that's what's going to really be the game changer here I I don't need a

the game changer here I I don't need a model today that tells me how to brush

model today that tells me how to brush my teeth if I'm at broadcom developing

my teeth if I'm at broadcom developing software right I need specialized models

software right I need specialized models and that lowers the comput ne uh needs

and that lowers the comput ne uh needs it lowers the energy consumption it

it lowers the energy consumption it lowers your carbon footprint AI is is

lowers your carbon footprint AI is is going to do a lot of really good and

going to do a lot of really good and it's you're going to see a broader

it's you're going to see a broader ecosystem as we go further outout okay

ecosystem as we go further outout okay Victor 10 years yeah I mean well first

Victor 10 years yeah I mean well first of all 10 years in AI is kind of like a

of all 10 years in AI is kind of like a hundred years of normal stuff so when

hundred years of normal stuff so when you have to realize that this is really

you have to realize that this is really futuristic projection but I I would just

futuristic projection but I I would just agree I mean I think if you're te just

agree I mean I think if you're te just generally you really are trying to have

generally you really are trying to have the positive outcomes where you're

the positive outcomes where you're mindful of some of the other things and

mindful of some of the other things and I do think it's going to change uh

I do think it's going to change uh everybody's lives on a daily basis um

everybody's lives on a daily basis um you know there's a I've been in

you know there's a I've been in Computing for a long time and there's a

Computing for a long time and there's a fundamental shift uh in Computing away

fundamental shift uh in Computing away from general purpose computes and much

from general purpose computes and much more towards acceler of some sort and I

more towards acceler of some sort and I also think it yes there will be very

also think it yes there will be very large foundational models but there will

large foundational models but there will also be much more mod models that you

also be much more mod models that you know in the bed space we're in like you

know in the bed space we're in like you know sensors intelligent sensors single

know sensors intelligent sensors single cameras you know let alone you know an

cameras you know let alone you know an edge Computing device um and there'll be

edge Computing device um and there'll be agents I I really liked also how uh

agents I I really liked also how uh Andrew talked about agents they'll be

Andrew talked about agents they'll be intelligent agents that are advocates

intelligent agents that are advocates for you right and I think they'll be um

for you right and I think they'll be um it's just really going to I hope that we

it's just really going to I hope that we will accelerate the outcomes in in

will accelerate the outcomes in in healthcare too because that's certainly

healthcare too because that's certainly whether you're in in it um I think all

whether you're in in it um I think all of us have you know have have in their

of us have you know have have in their personal life a touch point on

personal life a touch point on Healthcare so that's you know super um

Healthcare so that's you know super um important that we uh we get those

important that we uh we get those breakthroughs absolutely moer so I'm

breakthroughs absolutely moer so I'm going to say I don't know I don't think

going to say I don't know I don't think anyone knows uh because I think in the

anyone knows uh because I think in the last year it's been quite astonishing to

last year it's been quite astonishing to and to think about the next 10 years I

and to think about the next 10 years I think it's uh kind of unknowable but I I

think it's uh kind of unknowable but I I think there's two trends that we have to

think there's two trends that we have to look out for one is that um AI is a very

look out for one is that um AI is a very centralizing

centralizing Force you have more data you have more

Force you have more data you have more compute which gives you more data and

compute which gives you more data and more compute right you're able to

more compute right you're able to extract value You' be able to create

extract value You' be able to create your grow your team extract more

your grow your team extract more resources on the other hand one of the

resources on the other hand one of the really hopeful areas here is is around

really hopeful areas here is is around open source like without open source I'd

open source like without open source I'd be quite pessimistic uh so thank you for

be quite pessimistic uh so thank you for everyone you know contribute to open

everyone you know contribute to open source um so so maybe a lot of AI

source um so so maybe a lot of AI capabilities will be democratized so

capabilities will be democratized so that's kind of a hopeful hopeful uh

that's kind of a hopeful hopeful uh worldview all right blanch want to go

worldview all right blanch want to go next yes thank you so yeah I share

next yes thank you so yeah I share everything that has been said uh and I

everything that has been said uh and I truly believe that open source is going

truly believe that open source is going to create some more variety because

to create some more variety because everyone is able to to adapt open source

everyone is able to to adapt open source and to build on it so it's going to

and to build on it so it's going to Foster Innovation and and and open

Foster Innovation and and and open market for sure my vision is a world

market for sure my vision is a world where comput is is a is everywhere so

where comput is is a is everywhere so today it's difficult to get some

today it's difficult to get some European compute for example and and

European compute for example and and here I'm speaking about Europe I know

here I'm speaking about Europe I know I'm not even speaking about other areas

I'm not even speaking about other areas of the world so I think that in 10 years

of the world so I think that in 10 years I see a world where everyone has access

I see a world where everyone has access to AI everyone has access to computes

to AI everyone has access to computes and and we are Greener and better in the

and and we are Greener and better in the way we distribute compute and the way we

way we distribute compute and the way we use a variety of models that are smaller

use a variety of models that are smaller when the need is smaller and we I think

when the need is smaller and we I think we we we're going to be smarter in the

we we we're going to be smarter in the way we use the variety of AIS that will

way we use the variety of AIS that will are going to be available

are going to be available everywhere got it I hope so Alex final

everywhere got it I hope so Alex final words what's the yeah uh assuming AI

words what's the yeah uh assuming AI remains a transformational technology I

remains a transformational technology I it's an if I I don't take these things

it's an if I I don't take these things for granted

for granted uh it's my hope that events like this

uh it's my hope that events like this reflect uh you know an increased uh

reflect uh you know an increased uh Attunement to the importance of

Attunement to the importance of competition both for models themselves

competition both for models themselves but also in the supply chain for models

but also in the supply chain for models and I think we can hope that produces

and I think we can hope that produces positive results in competitive

positive results in competitive markets all right competitive markets

markets all right competitive markets five to 10 years we hope um thank you

five to 10 years we hope um thank you all thank you so much for the panelists

all thank you so much for the panelists thank you to all of you um I know I'm

thank you to all of you um I know I'm between you in lunch um so Jennifer's

between you in lunch um so Jennifer's probably going to orient you where to go

probably going to orient you where to go next yes yes thank you um we are going

next yes yes thank you um we are going to have lunch now out in the courtyard

to have lunch now out in the courtyard for about 45 minutes uh there's box

for about 45 minutes uh there's box lunch out there for you I encourage you

lunch out there for you I encourage you all to come back in 45 minutes because

all to come back in 45 minutes because we have a uh we're very grateful that we

we have a uh we're very grateful that we have recorded remarks from senator Amy

have recorded remarks from senator Amy kobitar who will be talking about her

kobitar who will be talking about her work uh to continue to promote

work uh to continue to promote competition in digital markets and AI in

competition in digital markets and AI in the Senate so please do come back for

the Senate so please do come back for those recorded remarks and about about

those recorded remarks and about about 15 in about 15 minutes in 45 minutes

15 in about 15 minutes in 45 minutes I'll give you more than 15 minutes uh

I'll give you more than 15 minutes uh and see you

seat all right um welcome back so for those of you who don't know me uh my

those of you who don't know me uh my name is jalo streer I'm the executive

name is jalo streer I'm the executive director of seer um we had a really

director of seer um we had a really great productive morning uh lots of

great productive morning uh lots of great discussions I really look forward

great discussions I really look forward to the second half of the conference uh

to the second half of the conference uh so thank you all for being here and I

so thank you all for being here and I also want to say thank you to our

also want to say thank you to our partners at the Department of Justice

partners at the Department of Justice and also at The Graduate School of

and also at The Graduate School of Business um so next we will hear from uh

Business um so next we will hear from uh some remarks by senator Amy kachar from

some remarks by senator Amy kachar from Minnesota I will not read her long list

Minnesota I will not read her long list of achievements uh but I just want to

of achievements uh but I just want to mention Senator clar is the first woman

mention Senator clar is the first woman elected to represent the state of

elected to represent the state of Minnesota in the US Senate and her 2021

Minnesota in the US Senate and her 2021 book antitrust was on the New York Times

book antitrust was on the New York Times best sellers list so let's hear from the

best sellers list so let's hear from the senator hello to everyone gathered for

senator hello to everyone gathered for today's important discussion on

today's important discussion on promoting competition in AI thank you to

promoting competition in AI thank you to everyone at Stanford and the justice

everyone at Stanford and the justice department who work to put this event

department who work to put this event together including my friend Assistant

together including my friend Assistant Attorney General Jonathan caner he's

Attorney General Jonathan caner he's just been a little bit busy lately like

just been a little bit busy lately like the work on Ticket Master I wish I could

the work on Ticket Master I wish I could be there with you in person as you know

be there with you in person as you know I actually wrote a book on

I actually wrote a book on competition and many of you may remember

competition and many of you may remember that during the presidential campaign I

that during the presidential campaign I had kind of a rivalry going with Pete

had kind of a rivalry going with Pete bootes well the first thing you should

bootes well the first thing you should know is we're actually good friends but

know is we're actually good friends but why did is this relevant to your topic

why did is this relevant to your topic today well when the campaign ended Pete

today well when the campaign ended Pete put out a New York Times bestselling

put out a New York Times bestselling book called trust not to be outdone

book called trust not to be outdone eight months later I put out my own New

eight months later I put out my own New York Times bestselling book for two

York Times bestselling book for two weeks called

weeks called antitrust it is no secret that our

antitrust it is no secret that our country has a competition problem that's

country has a competition problem that's why like you I'm working to restore

why like you I'm working to restore competition in our economy especially

competition in our economy especially when it comes to our digital markets

when it comes to our digital markets amid the rise of AI this work could not

amid the rise of AI this work could not be more important as we all know AI

be more important as we all know AI brings both opportunity and

brings both opportunity and uncertainty I think David Brooks put it

uncertainty I think David Brooks put it well when he wrote the people in I in in

well when he wrote the people in I in in AI seem to be experiencing radically

AI seem to be experiencing radically different brain States all at once I

different brain States all at once I found it incredibly hard to write about

found it incredibly hard to write about AI he said because it is literally

AI he said because it is literally unknowable whether this technology is

unknowable whether this technology is leading us to heaven or hell

leading us to heaven or hell his point is particularly resonant when

his point is particularly resonant when it comes to competition in AI the

it comes to competition in AI the emergence of any new technology can

emergence of any new technology can create enormous opportunities

create enormous opportunities opportunities when it comes to new

opportunities when it comes to new technologies I know Minnesota home of

technologies I know Minnesota home of Mayo Clinic we see enormous

Mayo Clinic we see enormous possibilities for curing diseases and

possibilities for curing diseases and treating diseases it also creates

treating diseases it also creates enormous opportunities for competition

enormous opportunities for competition new companies have seemingly spouted up

new companies have seemingly spouted up overnight firms no one had heard of a

overnight firms no one had heard of a few years ago like open Ai and anthropic

few years ago like open Ai and anthropic are becoming household names we've also

are becoming household names we've also seen a wave of smaller startups forming

seen a wave of smaller startups forming around AI Technologies with an estimated

around AI Technologies with an estimated more than 300 billion invested into over

more than 300 billion invested into over 25,000 Ai and machine learning startups

25,000 Ai and machine learning startups over the past three years but we may not

over the past three years but we may not necessarily be entering a golden age of

necessarily be entering a golden age of competition many of the most prominent

competition many of the most prominent AI startups are partner with the big

AI startups are partner with the big Tech incumbents rather than competing

Tech incumbents rather than competing with them of course business

with them of course business Partnerships Investments and even

Partnerships Investments and even Acquisitions are not necessarily

Acquisitions are not necessarily anti-competitive we all know that but we

anti-competitive we all know that but we can't make the same mistake with AI that

can't make the same mistake with AI that we did with prior Technologies by

we did with prior Technologies by allowing markets to consolidate and

allowing markets to consolidate and turning a blind eye to exclusionary

turning a blind eye to exclusionary conduct that's why we need to look out

conduct that's why we need to look out for anti-competitive behavior all the

for anti-competitive behavior all the way down the AI supply chain we must

way down the AI supply chain we must ensure the markets that are vital to the

ensure the markets that are vital to the safe and responsible development of AI

safe and responsible development of AI allow for the entry of new Innovative

allow for the entry of new Innovative startups that can compete the

startups that can compete the development of AI models requires

development of AI models requires enormous amounts of computing power that

enormous amounts of computing power that often only large firms can provide yet

often only large firms can provide yet cloud computing is largely controlled by

cloud computing is largely controlled by the very same firms developing the most

the very same firms developing the most advanced AI models as a result companies

advanced AI models as a result companies May preference proprietary AI models or

May preference proprietary AI models or products rather than supporting new or

products rather than supporting new or smaller firms terms even further down

smaller firms terms even further down the supply chain only one company

the supply chain only one company currently makes chips best suited for

currently makes chips best suited for the most advanced AI products of course

the most advanced AI products of course there's nothing wrong with making the

there's nothing wrong with making the best chips operating data centers or

best chips operating data centers or investing in startups but we must ensure

investing in startups but we must ensure that these markets are governed by

that these markets are governed by competitive market dynamics not Monopoly

competitive market dynamics not Monopoly power barriers to entry and

power barriers to entry and consolidation times of technological

consolidation times of technological transformation require critical thinking

transformation require critical thinking and Nuance to sure Innovation benefits

and Nuance to sure Innovation benefits consumers and businesses of all sizes

consumers and businesses of all sizes that's why discussions like this one are

that's why discussions like this one are so important it's also why I'm working

so important it's also why I'm working to ensure that our antitrust enforcers

to ensure that our antitrust enforcers have the resources and tools necessary

have the resources and tools necessary to carry out this vital task in many

to carry out this vital task in many cases these agencies even when you go

cases these agencies even when you go back to the time of Richard Nixon are

back to the time of Richard Nixon are shadows of their former C in terms of

shadows of their former C in terms of numbers of employees so last Congress we

numbers of employees so last Congress we passed my merger fee change legislation

passed my merger fee change legislation with Senator Chuck Grassley to give

with Senator Chuck Grassley to give enforcers at the FDC and at Justice the

enforcers at the FDC and at Justice the tools they need to keep a watchful eye

tools they need to keep a watchful eye over consolidation and anti-competitive

over consolidation and anti-competitive practices across the economy I also

practices across the economy I also continue to push Congress to pass my

continue to push Congress to pass my bipartisan Bill to prohibit online self

bipartisan Bill to prohibit online self preferencing by dominant Tech platforms

preferencing by dominant Tech platforms including AI platforms they spent the

including AI platforms they spent the tech companies about $300 million

tech companies about $300 million against a bill that's a minimum figure

against a bill that's a minimum figure uh in t tv ads I think at some point my

uh in t tv ads I think at some point my time will come I think at some point

time will come I think at some point when you look at what's going over the

when you look at what's going over the world we will come to an agreement on

world we will come to an agreement on some rules of the road when it comes to

some rules of the road when it comes to Tech

Tech competition to promote competition

competition to promote competition across all sectors of eon our economy I

across all sectors of eon our economy I also recently reinduced reintroduced my

also recently reinduced reintroduced my comprehensive legislation to empower

comprehensive legislation to empower enforcers to crack down on

enforcers to crack down on anti-competitive Acquisitions and

anti-competitive Acquisitions and exclusionary conduct that's a different

exclusionary conduct that's a different Bill yes we have a lot of work to do but

Bill yes we have a lot of work to do but as David Brooks wondered whether AI is

as David Brooks wondered whether AI is leading us to heaven or hell it's not

leading us to heaven or hell it's not out of our hands we don't just throw up

out of our hands we don't just throw up our hands like I'm afraid we did in the

our hands like I'm afraid we did in the past which has led to a lot of these

past which has led to a lot of these issues of what kids are exposed to in

issues of what kids are exposed to in the internet uh what's happening to our

the internet uh what's happening to our news organizations and democracy right

news organizations and democracy right now we have it in our power to put some

now we have it in our power to put some rules of the road in place we are not

rules of the road in place we are not just passive bystanders just few weeks

just passive bystanders just few weeks ago we passed three bipartisan bills out

ago we passed three bipartisan bills out of the rules committee that I chair to

of the rules committee that I chair to protect our elections from deceptive

protect our elections from deceptive uses of AI and that includes Banning

uses of AI and that includes Banning deep fakes a bill that I have with Josh

deep fakes a bill that I have with Josh Holly and Susan Collins that includes

Holly and Susan Collins that includes disclaimers on certain AI videos and

disclaimers on certain AI videos and Robo calls and the like otherwise people

Robo calls and the like otherwise people aren going to know if they're seeing the

aren going to know if they're seeing the candidate they love or the candidate

candidate they love or the candidate they don't like they won't know if

they don't like they won't know if they're real we cannot do that to our

they're real we cannot do that to our democracy and as a chair of of the

democracy and as a chair of of the competition policy and I trust in

competition policy and I trust in consumer rights subcommittee I'm

consumer rights subcommittee I'm fighting to ensure our markets remain

fighting to ensure our markets remain Fair open and contestable working

Fair open and contestable working together we can ensure that competition

together we can ensure that competition continues to drive innovation in

continues to drive innovation in America's economy for generations to

America's economy for generations to come I love Innovation I come from the

come I love Innovation I come from the state that gave the world everything

state that gave the world everything from the Post-it to the pacemaker and

from the Post-it to the pacemaker and certainly great great Innovations have

certainly great great Innovations have come out of Northern California but it

come out of Northern California but it is on all of us all of us to ensure that

is on all of us all of us to ensure that we see that next great product that next

we see that next great product that next gr Innovation some of the exciting

gr Innovation some of the exciting things we're seeing in AI but if we

things we're seeing in AI but if we don't put the guard rails in place that

don't put the guard rails in place that eventually is not going to work out for

eventually is not going to work out for people so I'm asking you to join us with

people so I'm asking you to join us with a business mind actually with a mind

a business mind actually with a mind that says we love competition we love

that says we love competition we love capitalism that's what's gotten us to

capitalism that's what's gotten us to where we are in America but we also know

where we are in America but we also know what Adam Smith warned our founding

what Adam Smith warned our founding father s watch out for the unbridled

father s watch out for the unbridled power of the army of monopolies if you

power of the army of monopolies if you don't put the guard rules in place you

don't put the guard rules in place you lose that Competitive Edge thank you

lose that Competitive Edge thank you enjoy today's discussion you got a lot

enjoy today's discussion you got a lot on your plate but I know you're up for

on your plate but I know you're up for it thanks

everyone okay so um now let's move on to the next panel uh our next panel will

the next panel uh our next panel will cover competition and innovation in

cover competition and innovation in Foundation models uh we will hear about

Foundation models uh we will hear about how companies are incorporating AI in

how companies are incorporating AI in their products and services and how

their products and services and how investors think about the value of AI

investors think about the value of AI for the world and the regulation in AI

for the world and the regulation in AI so this is a very complicated area very

so this is a very complicated area very important uh I'm really thrilled to

important uh I'm really thrilled to introduce our moderator uh professor

introduce our moderator uh professor mosam bayadi professor bayti is an

mosam bayadi professor bayti is an associate professor of operations

associate professor of operations information and Technology at Stanford

information and Technology at Stanford University Graduate School of Business

University Graduate School of Business uh so please join me in welcoming

uh so please join me in welcoming professor miad and uh the panel to the

professor miad and uh the panel to the [Applause]

welcome everyone and thank you for attending this panel uh where we explore

attending this panel uh where we explore competition in Foundation models and

competition in Foundation models and Beyond we also examine how companies

Beyond we also examine how companies employ AI uh barriers to

employ AI uh barriers to competition what drives investment in AI

competition what drives investment in AI technology

technology markets I'm thrilled to moderate this

markets I'm thrilled to moderate this panel with featuring such exceptional

panel with featuring such exceptional experts I will briefly introduce their

experts I will briefly introduce their names and current titles and you have

names and current titles and you have access to their full bio in the program

access to their full bio in the program so we have uh Karen Cen Chief data

so we have uh Karen Cen Chief data technology and inside officer from UK

technology and inside officer from UK competition and markets Authority or

competition and markets Authority or CMA uh Jes goind

CMA uh Jes goind deran senior vice president at

deran senior vice president at Salesforce AI uh and next speak next

Salesforce AI uh and next speak next speaker is wki ganison partner at menow

speaker is wki ganison partner at menow ventures and uh David George General

ventures and uh David George General partner from anden Horwitz or

partner from anden Horwitz or a16z so the structure of the panel is

a16z so the structure of the panel is we're going to start uh each panelist

we're going to start uh each panelist will give us introductory remarks about

will give us introductory remarks about their work and the topic of the panel

their work and the topic of the panel and then we'll follow by a series of

and then we'll follow by a series of questions that we have prepared and then

questions that we have prepared and then we'll open at the end for the audience's

we'll open at the end for the audience's questions

questions okay so we're going to start uh maybe

okay so we're going to start uh maybe with Karen uh I have uh a question about

with Karen uh I have uh a question about uh can you tell us about the work uh the

uh can you tell us about the work uh the CMA has been doing to study the

CMA has been doing to study the competition in the a St yeah thanks so

competition in the a St yeah thanks so much and uh really appreciate the

much and uh really appreciate the invitation to be here today obviously

invitation to be here today obviously such critical discussions and it's uh

such critical discussions and it's uh it's a nice opportunity to connect and

it's a nice opportunity to connect and to share a little bit about our work at

to share a little bit about our work at the CMA on Ai and and and recently also

the CMA on Ai and and and recently also AI Foundation models within that um so

AI Foundation models within that um so let me just introduce myself briefly in

let me just introduce myself briefly in the role and the focus if you like so I

the role and the focus if you like so I am the chief data technology and insight

am the chief data technology and insight officer at CMA I joined the CMA

officer at CMA I joined the CMA relatively recently so it's last year um

relatively recently so it's last year um I joined in a role there to lead the

I joined in a role there to lead the work of something called our data unit

work of something called our data unit data is an acronym there I so more about

data is an acronym there I so more about that in a moment um and really uh the

that in a moment um and really uh the remit has expanded quite a bit since

remit has expanded quite a bit since then and I'm now leading the cma's full

then and I'm now leading the cma's full portfolio of work on technology data and

portfolio of work on technology data and analytics um so that's all very exciting

analytics um so that's all very exciting for me um it's obviously it's an

for me um it's obviously it's an exciting time to serve in a role like

exciting time to serve in a role like this and I think actually the creation

this and I think actually the creation of a role like this is a real credit to

of a role like this is a real credit to the vision uh of agencies like the CMA

the vision uh of agencies like the CMA um so you know why have a role like this

um so you know why have a role like this well technology is reshaping markets and

well technology is reshaping markets and you know particularly these newer more

you know particularly these newer more emerging Technologies and also at the

emerging Technologies and also at the CMA we're taking on new powers um to

CMA we're taking on new powers um to ensure that we will be well placed to

ensure that we will be well placed to address some of the unique uh features

address some of the unique uh features and challenges of digital markets when

and challenges of digital markets when we look ahead and then in my role when I

we look ahead and then in my role when I think about all of that I see two broad

think about all of that I see two broad priorities um the first is that we

priorities um the first is that we really need then to understand um firm's

really need then to understand um firm's use of Technology uh all this data the

use of Technology uh all this data the emerging capabilities of things like Ai

emerging capabilities of things like Ai and then we need to back out the

and then we need to back out the implications for consumers and

implications for consumers and competition so that's the first priority

competition so that's the first priority the second actually is that we need then

the second actually is that we need then ourselves as an authority um also to

ourselves as an authority um also to harness all of the same advancing

harness all of the same advancing technology all that capability lots of

technology all that capability lots of uh large scale novel data so that we can

uh large scale novel data so that we can operate as efficiently and effectively

operate as efficiently and effectively as we can as an agency uh there's a huge

as we can as an agency uh there's a huge opportunity there so all of that

opportunity there so all of that requires quite a lot of technical

requires quite a lot of technical capability uh and the CMA has made a

capability uh and the CMA has made a significant investment in this space um

significant investment in this space um we've got a dedicated inhouse capability

we've got a dedicated inhouse capability this data unit I referred to stands for

this data unit I referred to stands for data technology analytics it was set up

data technology analytics it was set up about five years ago the focus back then

about five years ago the focus back then was data science and engineering but

was data science and engineering but it's really expanded a lot today when

it's really expanded a lot today when you look at that unit it's a very

you look at that unit it's a very interdisciplinary uh substantial unit it

interdisciplinary uh substantial unit it has skill set still in data science and

has skill set still in data science and engineering but also Behavioral Science

engineering but also Behavioral Science technology Insight we've got some

technology Insight we've got some digital forensics ecovery we've got a

digital forensics ecovery we've got a covert internet lab and we're adding

covert internet lab and we're adding some capability as well so we there we

some capability as well so we there we have uh in this unit a really

have uh in this unit a really interesting capability a strong

interesting capability a strong technical capability and we're using

technical capability and we're using that in many ways already across the

that in many ways already across the organization it's giving us quite a high

organization it's giving us quite a high return so some of the things we're able

return so some of the things we're able to do now are really better meet the

to do now are really better meet the demands of more complex cases including

demands of more complex cases including all of the major digital cases that my

all of the major digital cases that my teams work on we can scrutinize firm's

teams work on we can scrutinize firm's use of algorithm Ms including Emergen

use of algorithm Ms including Emergen emerging technology like AI uh and

emerging technology like AI uh and really work out the implications for

really work out the implications for consumers and competition and also our

consumers and competition and also our teams there you know often in

teams there you know often in partnership with other teams like our

partnership with other teams like our digital markets unit we drive proactive

digital markets unit we drive proactive thematic work on important topics like

thematic work on important topics like online Choice architecture topics like

online Choice architecture topics like uh Ai and and I think actually our

uh Ai and and I think actually our recent work brings me to our recent work

recent work brings me to our recent work on Foundation models which I think is a

on Foundation models which I think is a really good example of that and I don't

really good example of that and I don't know whether you want to keep going but

know whether you want to keep going but very be very happy of course to share a

very be very happy of course to share a little bit more in depth uh on that work

little bit more in depth uh on that work thank you we're going to come back with

thank you we're going to come back with the followup questions so I actually

the followup questions so I actually what happened I ended up asking you your

what happened I ended up asking you your first question but we can come back uh

first question but we can come back uh and uh so for the next speakers you

and uh so for the next speakers you prefer that I ask you the your first

prefer that I ask you the your first question as well or yeah okay so maybe

question as well or yeah okay so maybe uh we'll move to J and uh you can give

uh we'll move to J and uh you can give us some introductory remarks about your

us some introductory remarks about your work on the topic of the panel but the

work on the topic of the panel but the first question is can you start by

first question is can you start by sharing examples of the most exciting

sharing examples of the most exciting tools you're building with AI and how

tools you're building with AI and how they benefit customers and what is Sal

they benefit customers and what is Sal Force's philosophy on how to incorporate

Force's philosophy on how to incorporate AI into its products a great question

AI into its products a great question let me start by a quick introduction

let me start by a quick introduction first of all pleasure to be here uh my

first of all pleasure to be here uh my name is Jes I lead Salesforce uh AI

name is Jes I lead Salesforce uh AI engineering and science teams Salesforce

engineering and science teams Salesforce for those of you that don't know is an

for those of you that don't know is an Enterprise software company and our

Enterprise software company and our mission is to help customers uh find

mission is to help customers uh find their customers retain them and manage

their customers retain them and manage the customer relationship so that these

the customer relationship so that these customers large and small can focus on

customers large and small can focus on what they're great at which is building

what they're great at which is building products and services that Delight their

products and services that Delight their customers um my role in sales force is I

customers um my role in sales force is I lead the Salesforce AI organization uh

lead the Salesforce AI organization uh engineering and science teams and uh I'm

engineering and science teams and uh I'm responsible for uh building software

responsible for uh building software services that Infuse uh every aspect of

services that Infuse uh every aspect of AI within the flow of

AI within the flow of work um uh your question was toofold um

work um uh your question was toofold um let me start with the first one things

let me start with the first one things that I'm most excited about uh no

that I'm most excited about uh no surprise that are uh incredibly excited

surprise that are uh incredibly excited about um how generative AI is

about um how generative AI is transforming work uh and the ability to

transforming work uh and the ability to get work done better faster um uh my

get work done better faster um uh my team's currently working on um uh

team's currently working on um uh something called uh co-pilots which are

something called uh co-pilots which are in essence an assistant for every

in essence an assistant for every Persona um in you at work Salesforce uh

Persona um in you at work Salesforce uh we've been building software for over

we've been building software for over two decades now we understand quite

two decades now we understand quite deeply uh people that that uh run sales

deeply uh people that that uh run sales teams people that run service

teams people that run service organizations and we have a deep

organizations and we have a deep definition of what what it means to go

definition of what what it means to go get that job done uh incuding all of

get that job done uh incuding all of that into assistance being able to blend

that into assistance being able to blend generative and uh predictive AI

generative and uh predictive AI capabilities in a way that come together

capabilities in a way that come together um to create an assistant for every

um to create an assistant for every Persona that can make work easy is sort

Persona that can make work easy is sort of what my team is uh deeply focused on

of what my team is uh deeply focused on uh of course a lot of this uh employs

uh of course a lot of this uh employs large language models both from uh the

large language models both from uh the Frontier Model provider providers as

Frontier Model provider providers as well as open source models um uh all in

well as open source models um uh all in service of getting an assistant built um

service of getting an assistant built um how we do it was the second part of the

how we do it was the second part of the question um salesforce's AI stack we

question um salesforce's AI stack we call call it Einstein one platform um

call call it Einstein one platform um are one one core tenant of ours is open

are one one core tenant of ours is open and

and extensible uh this has always been true

extensible uh this has always been true uh this is where our customers take us

uh this is where our customers take us um the stack consists of a data layer it

um the stack consists of a data layer it consists of foundational model layer as

consists of foundational model layer as well as predictive models layer and

well as predictive models layer and Transformers that we run uh on top um on

Transformers that we run uh on top um on top of that is the co-pilot uh

top of that is the co-pilot uh orchestration and execution layer on top

orchestration and execution layer on top of that all the applications and

of that all the applications and assistance that we build at every layer

assistance that we build at every layer we allow extensibility which is uh to

we allow extensibility which is uh to say our customers can bring in a large

say our customers can bring in a large language model of their choice we love

language model of their choice we love that uh they can bring in their data

that uh they can bring in their data from any Lake we all that and we all

from any Lake we all that and we all customization um right at the app level

customization um right at the app level as well so in short thank thank you J uh

as well so in short thank thank you J uh I'm going to move to VY um so we're

I'm going to move to VY um so we're looking forward to hear about your work

looking forward to hear about your work and particularly what are successful AI

and particularly what are successful AI implementations you're seeing taking

implementations you're seeing taking place now uh what type of firms uh

place now uh what type of firms uh Market leaders or challengers uh are

Market leaders or challengers uh are having most

having most success thank you so much for having me

success thank you so much for having me it's very important conversation I think

it's very important conversation I think when you look at the history of

when you look at the history of Regulation now can't this success has

Regulation now can't this success has been mixed and I think it's been mixed

been mixed and I think it's been mixed because we have not had conversations

because we have not had conversations like this up front so I'm grateful we

like this up front so I'm grateful we are having this conversation it's a

are having this conversation it's a complicated issue a little bit about my

complicated issue a little bit about my background I'm a managing partner at men

background I'm a managing partner at men Ventures menal Ventures is one of the

Ventures menal Ventures is one of the oldest firms in silon Valley we are 46

oldest firms in silon Valley we are 46 years old which in the context of the

years old which in the context of the east coast is not or the UK not that old

east coast is not or the UK not that old but in Silicon Valley that's almost

but in Silicon Valley that's almost ancient um and we have been all in on AI

ancient um and we have been all in on AI so we are pivoted the firm to completely

so we are pivoted the firm to completely focus on AI it's not an area that's new

focus on AI it's not an area that's new to us we've been thinking about this for

to us we've been thinking about this for the last 15 years uh and because of that

the last 15 years uh and because of that focus on AI we are the largest or lead

focus on AI we are the largest or lead investors in foundational models like

investors in foundational models like anthropic outside of Google and Amazon

anthropic outside of Google and Amazon we the largest investor in anthropic

we the largest investor in anthropic Vector database companies like pine cone

Vector database companies like pine cone other Technologies like unstructured and

other Technologies like unstructured and a variety of app companies so that's

a variety of app companies so that's where our experience comes in I would

where our experience comes in I would say in terms of technologies that we are

say in terms of technologies that we are seeing a lot of traction in um I think

seeing a lot of traction in um I think chat jpt was kind of like the Netscape

chat jpt was kind of like the Netscape moment for AI for this audience looks

moment for AI for this audience looks like they actually know that reference

like they actually know that reference and I talked to my kids they don't know

and I talked to my kids they don't know what Netscape is so so when Netscape

what Netscape is so so when Netscape went out it's not like the internet

went out it's not like the internet didn't exist you know I was Art peret

didn't exist you know I was Art peret was there thanks to the government thank

was there thanks to the government thank you very much uh I used to internet when

you very much uh I used to internet when I was in college but when netk came out

I was in college but when netk came out it made the internet accessible to

it made the internet accessible to everyone including my mom and that

everyone including my mom and that actually

actually changed consumer adaptation ad option of

changed consumer adaptation ad option of the internet and I think that's what

the internet and I think that's what happens to AI it's not like AI is new

happens to AI it's not like AI is new you can go back there are papers in the

you can go back there are papers in the ' 70s the con convol neural Nets Minsky

' 70s the con convol neural Nets Minsky wrote about AI in the 70s so it's not

wrote about AI in the 70s so it's not like this is new but all of this is

like this is new but all of this is building to this moment and what chat CH

building to this moment and what chat CH did was go up and allowed everybody to

did was go up and allowed everybody to actually use it which then meant that we

actually use it which then meant that we have now crossed the chasm and people

have now crossed the chasm and people are adopting it and as we have as we get

are adopting it and as we have as we get to this adoption we got to think about

to this adoption we got to think about some other issues that it raises but

some other issues that it raises but when you come to the specific areas of

when you come to the specific areas of where we seeing the biggest adoptions I

where we seeing the biggest adoptions I would say in the Enterprise and large

would say in the Enterprise and large companies uh the Knowledge Management

companies uh the Knowledge Management Enterprise search is becoming a big area

Enterprise search is becoming a big area companies like sa gleen they're able to

companies like sa gleen they're able to collect all your information and answer

collect all your information and answer questions so anybody here who has worked

questions so anybody here who has worked in a large company knows they have a

in a large company knows they have a portal and you have to go and search for

portal and you have to go and search for documents while with generative AI

documents while with generative AI suddenly people are able to just ask

suddenly people are able to just ask questions a very common question people

questions a very common question people ask is what is a pan Le policy and can

ask is what is a pan Le policy and can that be extended

that be extended and previously you would have to go and

and previously you would have to go and look through documents to find out those

look through documents to find out those questions are answered by generative AI

questions are answered by generative AI so that's I would say like the number

so that's I would say like the number one use case number two use case is code

one use case number two use case is code generation and so this is where a

generation and so this is where a co-pilot is being sat next to your

co-pilot is being sat next to your development platform and it's allowing

development platform and it's allowing Engineers to being 30 to 40% more

Engineers to being 30 to 40% more productive those are the clear I would

productive those are the clear I would say use cases in the Enterprise on the

say use cases in the Enterprise on the consumer side we are seeing use cases

consumer side we are seeing use cases around conversational AI so I'm afraid

around conversational AI so I'm afraid if you didn't didn't like talking to

if you didn't didn't like talking to customer service before you're not going

customer service before you're not going to really love talking to the customer

to really love talking to the customer service agent AI because that's that's

service agent AI because that's that's your future you're not going to it's

your future you're not going to it's going to be very hard for you to get to

going to be very hard for you to get to a human but the good news is the AI is

a human but the good news is the AI is really good and almost makes you feel

really good and almost makes you feel like you're talking to a human so I'll

like you're talking to a human so I'll stop there and we'll go further thank

stop there and we'll go further thank you vinkki so uh David uh one a6z has

you vinkki so uh David uh one a6z has written extensively about little Tech uh

written extensively about little Tech uh you work closely with a large number of

you work closely with a large number of early stage companies what is a6c seeing

early stage companies what is a6c seeing in how the AI Market is developing and

in how the AI Market is developing and what are things that your companies are

what are things that your companies are excited about and what are they

excited about and what are they concerned about yeah so thanks for

concerned about yeah so thanks for having me here um uh vinkki I I for one

having me here um uh vinkki I I for one I'm very excited about uh customer

I'm very excited about uh customer support just with machines uh it can't

support just with machines uh it can't get much worse uh I think it can only

get much worse uh I think it can only get better um and you know no matter how

get better um and you know no matter how much you yell at an AI it's it's

much you yell at an AI it's it's infinitely patient um you know it won't

infinitely patient um you know it won't come back at you uh you know too badly

come back at you uh you know too badly um so uh just quick background on myself

um so uh just quick background on myself um I run what we call our growth fund

um I run what we call our growth fund which is the fund at Andre and Horowitz

which is the fund at Andre and Horowitz that invests in um the most mature

that invests in um the most mature companies that we invest in so than you

companies that we invest in so than you know companies that have sort of found

know companies that have sort of found product Market fit um and have some

product Market fit um and have some traction uh we at the firm are very

traction uh we at the firm are very large investors in AI we have been for a

large investors in AI we have been for a long time similar to vinkki

long time similar to vinkki um we're the largest or very large

um we're the largest or very large investors in uh leading Foundation

investors in uh leading Foundation models in uh language um both closed

models in uh language um both closed source and open source um in voice uh in

source and open source um in voice uh in image in video um in music um I'm sure

image in video um in music um I'm sure I'm missing some other categories but

I'm missing some other categories but all the main categories where there our

all the main categories where there our foundation models um and we're

foundation models um and we're astonished uh at what this technology

astonished uh at what this technology can do

can do I um I agreed with a lot of what Senator

I um I agreed with a lot of what Senator kachar said actually uh which might be

kachar said actually uh which might be somewhat surprising uh to people for a

somewhat surprising uh to people for a POS person in my position um one thing

POS person in my position um one thing that I did not agree with uh which I

that I did not agree with uh which I know she was quoting is is AI going to

know she was quoting is is AI going to be heaven or hell um I think taglines

be heaven or hell um I think taglines like that um you know serve really the

like that um you know serve really the only purpose uh is to strike fear in us

only purpose uh is to strike fear in us and I think would us to act potentially

and I think would us to act potentially hastily and create regulatory structures

hastily and create regulatory structures that actually could do more harm than

that actually could do more harm than good um and I'll talk a little bit about

good um and I'll talk a little bit about what I mean by that um with some

what I mean by that um with some examples um I think Professor you asked

examples um I think Professor you asked me about uh little Tech we came up with

me about uh little Tech we came up with this tagline at the firm uh to refer to

this tagline at the firm uh to refer to what our interest is so we're Venture

what our interest is so we're Venture investors obviously so we have a

investors obviously so we have a financial interest but we also think

financial interest but we also think this is what's best for society um which

this is what's best for society um which is to to enable little Tech or small

is to to enable little Tech or small businesses uh to have a fair shot to

businesses uh to have a fair shot to fight against big Tech um and one of the

fight against big Tech um and one of the things that we are seeing with uh a lot

things that we are seeing with uh a lot of the discussion around policy at this

of the discussion around policy at this point is um big Tech that is extremely

point is um big Tech that is extremely capable and extremely motivated so you

capable and extremely motivated so you know there's all these stories about

know there's all these stories about when the mobile phone came out um and

when the mobile phone came out um and you know Microsoft leadership sort of

you know Microsoft leadership sort of dismissing it and saying oh this is a a

dismissing it and saying oh this is a a silly little thing you know it's why

silly little thing you know it's why would you want that um and totally

would you want that um and totally missing the boat uh I think the big tech

missing the boat uh I think the big tech companies right now are very effective

companies right now are very effective they have extremely smart people uh both

they have extremely smart people uh both technologically uh and from a business

technologically uh and from a business and Regulatory standpoint uh and they

and Regulatory standpoint uh and they are throwing their weight around in a

are throwing their weight around in a way that we haven't seen in previous

way that we haven't seen in previous Cycles um so when the big tech companies

Cycles um so when the big tech companies come and try to make proposals often

come and try to make proposals often it's around safetyism or you know things

it's around safetyism or you know things like that um understand that it is

like that um understand that it is likely to be very self-interested

likely to be very self-interested and it's a blatant attempt in a lot of

and it's a blatant attempt in a lot of cases at regulatory capture and so our

cases at regulatory capture and so our attempts at backing and representing

attempts at backing and representing little Tech uh in the market uh sort of

little Tech uh in the market uh sort of revolve around that as the big idea and

revolve around that as the big idea and so a couple of the things that we are

so a couple of the things that we are most focused on um to that end um one is

most focused on um to that end um one is we have a great framework of rules and

we have a great framework of rules and regulations and laws that we can enforce

regulations and laws that we can enforce uh today um I I get the the idea behind

uh today um I I get the the idea behind creating new agencies and new compliance

creating new agencies and new compliance requirements but just understand that

requirements but just understand that every time we go down the path of one of

every time we go down the path of one of those it is competitively favoring big

those it is competitively favoring big Tech over little Tech and it's likely to

Tech over little Tech and it's likely to entrench them relative to startups who

entrench them relative to startups who are trying to compete but don't have the

are trying to compete but don't have the resources to invest a lot in compliance

resources to invest a lot in compliance purposes um the second big thing which I

purposes um the second big thing which I know we'll talk about in much more

know we'll talk about in much more detail and I know you've touched on

detail and I know you've touched on during the sessions already um is open

during the sessions already um is open source we believe it is AB absolutely

source we believe it is AB absolutely critical that we invest aggressively

critical that we invest aggressively behind open source um there are two main

behind open source um there are two main reasons why um one is we believe that

reasons why um one is we believe that it's our best way of staying ahead in

it's our best way of staying ahead in the worldwide race in AI um China is

the worldwide race in AI um China is moving very fast our other adversaries

moving very fast our other adversaries are moving very fast uh and we think

are moving very fast uh and we think having more eyes on things more

having more eyes on things more transparency is the most effective way

transparency is the most effective way for us to be maximally competitive uh as

for us to be maximally competitive uh as a country and and as a West um the

a country and and as a West um the second part is it's the most effective

second part is it's the most effective way to combat potential lock in with the

way to combat potential lock in with the big Tech players themselves Capital

big Tech players themselves Capital requirements are High um you know

requirements are High um you know getting getting products built are is

getting getting products built are is very difficult in this space open source

very difficult in this space open source allows other people to have access to

allows other people to have access to this technology with transparency around

this technology with transparency around what they're working with which big Tech

what they're working with which big Tech close model things will not provide um

close model things will not provide um so you know if I could sort of impart

so you know if I could sort of impart two big things uh as as it relates to

two big things uh as as it relates to little Tech um it's don't let big Tech

little Tech um it's don't let big Tech set the rules because those are likely

set the rules because those are likely to be

to be self-interested um and secondly we

self-interested um and secondly we should go much much harder on open

should go much much harder on open source and support it uh rather than try

source and support it uh rather than try and restrict it thank you David uh we're

and restrict it thank you David uh we're going to Karen come back to the topics

going to Karen come back to the topics you brought up so specifically you

you brought up so specifically you talked about uh some principles you

talked about uh some principles you identified and uh can you tell us more

identified and uh can you tell us more about the specifically key competition

about the specifically key competition risks identified and how your principles

risks identified and how your principles address these uh and how this is shaping

address these uh and how this is shaping cma's work and action in the space yeah

cma's work and action in the space yeah thanks um so we you know I mentioned our

thanks um so we you know I mentioned our work on algorithms and AI we've been

work on algorithms and AI we've been doing that for some time um recently you

doing that for some time um recently you know you'll see that a lot of our work

know you'll see that a lot of our work has been on Foundation models

has been on Foundation models specifically you know a word on why that

specifically you know a word on why that is um you know obviously I won't

is um you know obviously I won't rehearse uh a lot of what we've heard at

rehearse uh a lot of what we've heard at this excellent uh event already

this excellent uh event already these markets have huge potential

these markets have huge potential promise in lots of different ways right

promise in lots of different ways right through to um I think we heard Andrew un

through to um I think we heard Andrew un earlier talking about uh the some of the

earlier talking about uh the some of the possibilities around agentic systems so

possibilities around agentic systems so potentially a lot of Promise potentially

potentially a lot of Promise potentially a lot at stake of course a lot of

a lot at stake of course a lot of uncertainty as well and nobody really

uncertainty as well and nobody really has a a crystal ball here it's a little

has a a crystal ball here it's a little bit difficult to see quite how

bit difficult to see quite how transformative and and the shape of that

transformative and and the shape of that uh here in advance um but clearly

uh here in advance um but clearly considerations and I think when we think

considerations and I think when we think about our role as the CMA we feel a real

about our role as the CMA we feel a real responsibility to be using the full

responsibility to be using the full range of our powers to ensure that these

range of our powers to ensure that these promising Emerging Markets are

promising Emerging Markets are underpinned by Fair open effective

underpinned by Fair open effective competition and also I should say strong

competition and also I should say strong uh consumer protection so that was the

uh consumer protection so that was the motivation for us in May uh last year

motivation for us in May uh last year opening a an initial review and initial

opening a an initial review and initial period of work uh on Foundation models

period of work uh on Foundation models diving into this space a little and

diving into this space a little and looking at considerations across the

looking at considerations across the stack we published our first report back

stack we published our first report back in September um and you know what were

in September um and you know what were we doing there well really uh you know

we doing there well really uh you know we were approaching this with curiosity

we were approaching this with curiosity we were looking to build and share some

we were looking to build and share some early understanding um you know what uh

early understanding um you know what uh what are these Technologies how are they

what are these Technologies how are they being developed and deployed out in

being developed and deployed out in markets um what uh are what is the

markets um what uh are what is the future outlook what are some of the uh

future outlook what are some of the uh you know potentially evolutionary paths

you know potentially evolutionary paths here and what are the considerations

here and what are the considerations that we see emerging uh in this space

that we see emerging uh in this space for consumers and competition and we

for consumers and competition and we noted the many potential uh benefits of

noted the many potential uh benefits of this technology but we did also identify

this technology but we did also identify a risk that these markets could develop

a risk that these markets could develop in a way that would harm uh competition

in a way that would harm uh competition and consumers as well um and we you

and consumers as well um and we you mentioned our principles we laid out uh

mentioned our principles we laid out uh a set of draft AI principles to if you

a set of draft AI principles to if you like to guide this Market as it develops

like to guide this Market as it develops um and I'll come back to those in a

um and I'll come back to those in a moment and then we entered a period

moment and then we entered a period since September of really monitoring the

since September of really monitoring the many developments uh in this very active

many developments uh in this very active ecosystem very closely um it's been

ecosystem very closely um it's been really interesting we've engaged a lot

really interesting we've engaged a lot with a wide range of stakeholders across

with a wide range of stakeholders across industry Civil Society uh fellow

industry Civil Society uh fellow regulators and agencies um and we little

regulators and agencies um and we little Tech and big Tech um and we have taken

Tech and big Tech um and we have taken feedback on our principles and our work

feedback on our principles and our work um and you know when you fast forward to

um and you know when you fast forward to um to about a month ago you'll see that

um to about a month ago you'll see that we published a further update report we

we published a further update report we kept our analysis going and you know

kept our analysis going and you know what are we doing in this report well uh

what are we doing in this report well uh one thing we do is take stock of the

one thing we do is take stock of the developments we've been seeing the rise

developments we've been seeing the rise in use and experimentation by consumers

in use and experimentation by consumers with these models and outputs from these

with these models and outputs from these models um a a growth in the number of

models um a a growth in the number of models their

models their capability uh distribution and roots to

capability uh distribution and roots to Market um and an interesting and quite

Market um and an interesting and quite striking drum beat generally of

striking drum beat generally of innovation industry activity industry

innovation industry activity industry announcements U many relating to model

announcements U many relating to model launchers some relating to Partnerships

launchers some relating to Partnerships and investments in this space um so we

and investments in this space um so we took stock of these developments and

took stock of these developments and again you know I think we continued to

again you know I think we continued to note um the many potential benefits that

note um the many potential benefits that can occr from this kind of Technology

can occr from this kind of Technology out there for society and the economy

out there for society and the economy but we did also um identify or make a

but we did also um identify or make a key observation that it's possible to

key observation that it's possible to see that the largest most established

see that the largest most established technology firms are very present um

technology firms are very present um across the value chain and this is

across the value chain and this is manifesting in a few ways um this is

manifesting in a few ways um this is through vertical integration it's also

through vertical integration it's also through um uh a number of Partnerships

through um uh a number of Partnerships and Investments and we identified

and Investments and we identified actually a set of

actually a set of interconnected uh an interconnected set

interconnected uh an interconnected set of Investments involving the biggest uh

of Investments involving the biggest uh Tech firms I think there were over 90

Tech firms I think there were over 90 that we identified and of course this

that we identified and of course this fast moving space now I should say that

fast moving space now I should say that um obviously we do think that the

um obviously we do think that the largest technology firms can play an

largest technology firms can play an important role in this market have an

important role in this market have an important role to play when you think

important role to play when you think about uh the there were mentioned a

about uh the there were mentioned a moment ago of the the wealth of

moment ago of the the wealth of expertise and the resources in a market

expertise and the resources in a market like this with some of the fundamentals

like this with some of the fundamentals and um and and so there could be an

and um and and so there could be an important role uh for big dech to play

important role uh for big dech to play but we do know that digital markets can

but we do know that digital markets can tip very quickly to more of a winner

tip very quickly to more of a winner takes all scenario so it is really

takes all scenario so it is really essential to be uh Vigilant in a market

essential to be uh Vigilant in a market like this and we set out three interl uh

like this and we set out three interl uh key risks to uh Fair open effective

key risks to uh Fair open effective competition if there's a if there's time

competition if there's a if there's time I might touch on those very briefly in a

I might touch on those very briefly in a moment and tell you about some of our

moment and tell you about some of our work and actions um on the principles

work and actions um on the principles you know the principles that we um then

you know the principles that we um then confirmed last month in our report they

confirmed last month in our report they are access um diversity Choice Fair

are access um diversity Choice Fair dealing transparency and accountability

dealing transparency and accountability very briefly there with access uh this

very briefly there with access uh this is about access to the critical inputs

is about access to the critical inputs to develop these models the data the um

to develop these models the data the um compute the expertise um with diversity

compute the expertise um with diversity we're looking for a sustained variety of

we're looking for a sustained variety of of models and model types and we heard a

of models and model types and we heard a little bit about uh types of models as

little bit about uh types of models as well with Choice it's really about

well with Choice it's really about ensuring that businesses and consumers

ensuring that businesses and consumers are able and empowered to make uh

are able and empowered to make uh choices in relation to their use of

choices in relation to their use of these models Fair dealing would mean no

these models Fair dealing would mean no um uh anti competitive bundling tying or

um uh anti competitive bundling tying or self- preferencing and then transparency

self- preferencing and then transparency and accountability are really uh key as

and accountability are really uh key as well including for Consumer Protection

well including for Consumer Protection so with transparency it's about really

so with transparency it's about really making sure that the risks and the

making sure that the risks and the limitations of these models can be

limitations of these models can be understood including right through to

understood including right through to Consumers who will need to make these

Consumers who will need to make these decisions about their use and then

decisions about their use and then accountability we're really looking for

accountability we're really looking for that accountability across the entire

that accountability across the entire value chain so that's um the developers

value chain so that's um the developers Upstream right through to the deployers

Upstream right through to the deployers um and all of those involved in the

um and all of those involved in the value chain the accountability for the

value chain the accountability for the outputs and if there's a moment Mo I

outputs and if there's a moment Mo I might just mention these three uh interl

might just mention these three uh interl um key risks to competition so the first

um key risks to competition so the first is that the um the risk that um

is that the um the risk that um incumbent firms powerful firms with

incumbent firms powerful firms with control over critical a um critical um

control over critical a um critical um inputs for model development May

inputs for model development May constrain or materially uh restrict

constrain or materially uh restrict access to the in order to Shield

access to the in order to Shield themselves from

themselves from competition uh in terms of some of the

competition uh in terms of some of the uh you know obviously they're thinking

uh you know obviously they're thinking about our principles so access is a very

about our principles so access is a very important one I think also diversity uh

important one I think also diversity uh there too and in terms of some of the

there too and in terms of some of the actions that we're taking in this space

actions that we're taking in this space I'll mention our Cloud Market

I'll mention our Cloud Market investigation which is ongoing uh over

investigation which is ongoing uh over in the UK um and there we are looking at

in the UK um and there we are looking at competition in the cloud market and

competition in the cloud market and we're also going to be including a

we're also going to be including a forward-looking assessment of of the

forward-looking assessment of of the foundation models market and the impact

foundation models market and the impact that that could have on competition in

that that could have on competition in Cloud um we are looking at relevant

Cloud um we are looking at relevant Partnerships we are um looking at the AI

Partnerships we are um looking at the AI chips market and some of the

chips market and some of the considerations there for the foundation

considerations there for the foundation models ecosystem and I mentioned at the

models ecosystem and I mentioned at the start as well that we're taking on new

start as well that we're taking on new powers to enable us better to um to

powers to enable us better to um to ensure strong competition and consumer

ensure strong competition and consumer protection in digital markets and we

protection in digital markets and we will be going through an exercise of

will be going through an exercise of prioritizing digital activities areas to

prioritizing digital activities areas to focus on and of course um there'll be

focus on and of course um there'll be considerations there for our

considerations there for our prioritization and we haven't should add

prioritization and we haven't should add that we haven't taken any uh provisional

that we haven't taken any uh provisional decisions yet and everything will be

decisions yet and everything will be subject to full consideration and and

subject to full consideration and and investigation um the second risk

investigation um the second risk thinking now a little bit more

thinking now a little bit more Downstream is that then powerful

Downstream is that then powerful incumbents with with powerful positions

incumbents with with powerful positions in Downstream markets that are important

in Downstream markets that are important for the deployment of these models could

for the deployment of these models could distort choice in those markets and

distort choice in those markets and undermine competition in those markets

undermine competition in those markets and so when you look at Foundation

and so when you look at Foundation models and you think how they might be

models and you think how they might be deployed um you know what comes to mind

deployed um you know what comes to mind uh are some digital activities that

uh are some digital activities that might serve as critical access points or

might serve as critical access points or roots to Market so I'll mention mobile

roots to Market so I'll mention mobile ecosystems um search and productivity

ecosystems um search and productivity software as examples um and here too I

software as examples um and here too I think there are going to be important

think there are going to be important considerations for us as we take on

considerations for us as we take on these new powers and think about our

these new powers and think about our prioritization and then the final risk

prioritization and then the final risk to open effective competition I

to open effective competition I mentioned the Partnerships and the

mentioned the Partnerships and the investments in this space so a risk is

investments in this space so a risk is that um Partnerships involving key

that um Partnerships involving key players in this space and perhaps

players in this space and perhaps players with both with power both

players with both with power both upstream and downstream in these markets

upstream and downstream in these markets could inhibit Fair effective competition

could inhibit Fair effective competition and there we are using our we're

and there we are using our we're monitoring actively the Partnerships and

monitoring actively the Partnerships and Investments we are using our merger

Investments we are using our merger review powers to take a closer look at

review powers to take a closer look at relevant Partnerships uh and Investments

relevant Partnerships uh and Investments and everything needs to be considered on

and everything needs to be considered on its merits we know that Partnerships um

its merits we know that Partnerships um can play an important role in markets

can play an important role in markets like this they may be important for

like this they may be important for independent developers accessing

independent developers accessing critical inputs and getting to Market um

critical inputs and getting to Market um but nonetheless uh you know it's also

but nonetheless uh you know it's also true that these Partnerships and

true that these Partnerships and Investments the arrangements can

Investments the arrangements can sometimes be a little complex and opaque

sometimes be a little complex and opaque to understand and it is important to

to understand and it is important to ensure that strong competition is upheld

ensure that strong competition is upheld and so we are using our powers to take a

and so we are using our powers to take a look to provide some to gain some

look to provide some to gain some clarity over what could fall in the

clarity over what could fall in the scope of our merger powers and where

scope of our merger powers and where concerns about competition may arise and

concerns about competition may arise and we are looking really to provide that

we are looking really to provide that Clarity back to the market and we think

Clarity back to the market and we think that this CL ity will be in the

that this CL ity will be in the interests of firms as well thank you

interests of firms as well thank you Karen we're going to actually take the

Karen we're going to actually take the topic of partnership and go to J uh

topic of partnership and go to J uh there has been a number of such

there has been a number of such Partnerships between specifically llm

Partnerships between specifically llm providers and major public Cloud

providers and major public Cloud providers uh over last year do you have

providers uh over last year do you have any concern about how certain terms are

any concern about how certain terms are in this Partnerships like exclusivity

in this Partnerships like exclusivity type Arrangements could affect

type Arrangements could affect competition um that's a good question I

competition um that's a good question I think like I'm going to say something

think like I'm going to say something sort of general and then maybe uh follow

sort of general and then maybe uh follow up with a more specific example of of

up with a more specific example of of how that might not be the right thing um

how that might not be the right thing um certainly there is exclusive

certainly there is exclusive Arrangements between hyperscalers and

Arrangements between hyperscalers and large language model providers that we

large language model providers that we that we see and broadly it feels like

that we see and broadly it feels like we're very early in the cycle right now

we're very early in the cycle right now um I think um the frontier large

um I think um the frontier large language model providers are probably in

language model providers are probably in a step one or step two of a multi-step

a step one or step two of a multi-step Journey so it feels too early to be

Journey so it feels too early to be having such a degree of exclusivity not

having such a degree of exclusivity not just because of um how sophisticated

just because of um how sophisticated these models are today but the

these models are today but the trajectory of growth and where they're

trajectory of growth and where they're going to be so getting exclusive this

going to be so getting exclusive this early shuts down some future paths which

early shuts down some future paths which is um which is problematic and

is um which is problematic and definitely limiting both for the said

definitely limiting both for the said large language mods and for customers uh

large language mods and for customers uh specifically how this impacts us um uh

specifically how this impacts us um uh and builders in general um is often our

and builders in general um is often our customers ask us for choice right choice

customers ask us for choice right choice in the hyperscalers that they that they

in the hyperscalers that they that they want to work with and we give them that

want to work with and we give them that choice the stack that's built on um on

choice the stack that's built on um on any company Salesforce no exception uh

any company Salesforce no exception uh you know enables us to offer those

you know enables us to offer those services on any hyperscaler now Along

services on any hyperscaler now Along Comes A Foundation model that's

Comes A Foundation model that's available only on one hyperscaler that

available only on one hyperscaler that limits the choice that we are able to

limits the choice that we are able to offer to our customers right and I think

offer to our customers right and I think that is not just limiting from a

that is not just limiting from a customer persp perspective that's also

customer persp perspective that's also limiting from Innovation perspective uh

limiting from Innovation perspective uh the reason for that

the reason for that is the foundational models are

is the foundational models are foundational for a reason there are

foundational for a reason there are low-level building blocks uh that entire

low-level building blocks uh that entire Stacks are going to get built on so if

Stacks are going to get built on so if you have exclusive relationship between

you have exclusive relationship between infrastructure providers and the first

infrastructure providers and the first layer on top of that infrastructure prod

layer on top of that infrastructure prod which is a foundational layer you can

which is a foundational layer you can you can just see where that's going

you can just see where that's going right you can start to build these

right you can start to build these vertical Stacks that are entirely

vertical Stacks that are entirely exclusive

exclusive um so I'd say like those are two reasons

um so I'd say like those are two reasons one in general feels too early to be

one in general feels too early to be even thinking of exclusivity because you

even thinking of exclusivity because you don't want to morage a future um and

don't want to morage a future um and then of course from the perspective of

then of course from the perspective of choice and these vertical Stacks forming

choice and these vertical Stacks forming we don't want them to be that vertical

we don't want them to be that vertical right away can you give an example of

right away can you give an example of like those exclusive potential

like those exclusive potential exclusivity rights that could be yeah um

exclusivity rights that could be yeah um you know you can pick the world's best

you know you can pick the world's best um uh uh you know llm model that is soon

um uh uh you know llm model that is soon to be AGI is what we hear and you can

to be AGI is what we hear and you can pick uh you know

pick uh you know hyperscaler that um that it runs on so

hyperscaler that um that it runs on so if you want access to that large

if you want access to that large language model that is running in a data

language model that is running in a data center in let's say uh Singapore you

center in let's say uh Singapore you don't have much choice in the US you

don't have much choice in the US you might have a lot of choice in emia you

might have a lot of choice in emia you might have a lot of choice but if you

might have a lot of choice but if you want for for gdpr purposes you want to

want for for gdpr purposes you want to run that on in a certain region there's

run that on in a certain region there's only one game in town and that's because

only one game in town and that's because of that relationship I now I think

of that relationship I now I think that'll change because it's limiting

that'll change because it's limiting like I said both from a competive

like I said both from a competive perspective as well as from a um you

perspective as well as from a um you know commercial perspective so I think

know commercial perspective so I think that that will and should

that that will and should change thank you uh so the next question

change thank you uh so the next question is for uh I'm going to move to the

is for uh I'm going to move to the investment side VY do you think

investment side VY do you think regulation in AI is

regulation in AI is inevitable and if so what are some key

inevitable and if so what are some key principles that would promote

principles that would promote competitiveness particularly from the

competitiveness particularly from the perspective of

perspective of investment thank you and lots Karen I

investment thank you and lots Karen I love how you walked out that was

love how you walked out that was actually very comprehensive I don't know

actually very comprehensive I don't know if I caught everything but I definitely

if I caught everything but I definitely took a note to say I got to go and look

took a note to say I got to go and look through it um I think regulation in AI

through it um I think regulation in AI is inevitable Eternal vigilance is a

is inevitable Eternal vigilance is a price of Liberty and the only way you

price of Liberty and the only way you can be vigilant is you got to have now

can be vigilant is you got to have now the challenge is how do you regulate in

the challenge is how do you regulate in a way that is pro-innovation pro

a way that is pro-innovation pro competition and pro-america or Pro West

competition and pro-america or Pro West and I think this is not traditionally

and I think this is not traditionally being an issue when it comes to

being an issue when it comes to regulatory regimes is to think about the

regulatory regimes is to think about the geopolitical aspect but I think it's got

geopolitical aspect but I think it's got to be a very important element of how we

to be a very important element of how we think through so it can't just be about

think through so it can't just be about consumer Choice also got to be how do we

consumer Choice also got to be how do we make sure we don't handicap because we

make sure we don't handicap because we are in a geopolitical race and not

are in a geopolitical race and not everyone has the same rules and if we

everyone has the same rules and if we were to create a set of rules that

were to create a set of rules that handicap companies in the west I I think

handicap companies in the west I I think in this race that is a big deal so I

in this race that is a big deal so I would urge Regulators to think not just

would urge Regulators to think not just the context of consumer Choice which is

the context of consumer Choice which is important safety which is also important

important safety which is also important but how do you also make sure we don't

but how do you also make sure we don't handicap companies in the west because

handicap companies in the west because regardless of how much you might like or

regardless of how much you might like or dislike uh the US government snooping on

dislike uh the US government snooping on you trust me you're not going to allow

you trust me you're not going to allow the CCP snooping on you and whatever you

the CCP snooping on you and whatever you think about what our our companies do uh

think about what our our companies do uh companies in China are going to be much

companies in China are going to be much worse with your data so so to me how do

worse with your data so so to me how do we find that balance

we find that balance uh super important I think David you

uh super important I think David you mentioned something about you know big

mentioned something about you know big Tech and little Tech and I I would say

Tech and little Tech and I I would say yeah the problem with regulatory regimes

yeah the problem with regulatory regimes is just that you're going to be

is just that you're going to be inundated with lobbyists from Big Tech

inundated with lobbyists from Big Tech because they have lots of money they

because they have lots of money they they are successful businesses um we

they are successful businesses um we live in a democracy a political process

live in a democracy a political process so you just have to know the people

so you just have to know the people you're going to hear from and the people

you're going to hear from and the people you're going to spend a lot of time

you're going to spend a lot of time massaging our elected representatives

massaging our elected representatives are going to be people from Big teac

are going to be people from Big teac because it's in the agenda uh I would

because it's in the agenda uh I would say

say that to me you also have to think about

that to me you also have to think about timing there's a little benefit to wait

timing there's a little benefit to wait and watch and I know sometimes that's

and watch and I know sometimes that's not a popular opinion but like you want

not a popular opinion but like you want to have things evolve if you go too

to have things evolve if you go too early in your regulatory regime what

early in your regulatory regime what happens is you compress Innovation I

happens is you compress Innovation I mean again these are all complicated

mean again these are all complicated issues in which you can think but like I

issues in which you can think but like I think the early part of the internet

think the early part of the internet things like section 230 did allow the

things like section 230 did allow the internet to flourish now don't get me

internet to flourish now don't get me wrong there were issues with it and we

wrong there were issues with it and we will have to fix it but making sure

will have to fix it but making sure things are open in the

things are open in the beginning I think that's a good job so

beginning I think that's a good job so so coming back to like how do we make

so coming back to like how do we make sure is is really think about okay every

sure is is really think about okay every rule you implement the cost of

rule you implement the cost of conforming to the rule is significantly

conforming to the rule is significantly higher for smaller companies than they

higher for smaller companies than they are for large companies and so the

are for large companies and so the Regulatory and compliance burden right

Regulatory and compliance burden right good example is just what did sban soy

good example is just what did sban soy do which came out of you know what

do which came out of you know what happened in 2001 it really made public

happened in 2001 it really made public markets inaccessible for a lot of small

markets inaccessible for a lot of small companies well-intended rules can have

companies well-intended rules can have unintended consequences so I would just

unintended consequences so I would just really urge um regulatory authorities to

really urge um regulatory authorities to like take their time here think through

like take their time here think through make sure that you reach out to voices

make sure that you reach out to voices that may not be and I think a critical

that may not be and I think a critical voice that we should think about is also

voice that we should think about is also Academia right which which I think would

Academia right which which I think would have a more nonbiased or

have a more nonbiased or non-economically driven View and I think

non-economically driven View and I think one of the challenges I think I really

one of the challenges I think I really see is that Academia in academic

see is that Academia in academic institutions need to be funded to have

institutions need to be funded to have access to the best Technologies I think

access to the best Technologies I think I was listening to F Fe Lee talking

I was listening to F Fe Lee talking about how you know she doesn't have

about how you know she doesn't have access to gpus because it's really

access to gpus because it's really expensive like how is how are our

expensive like how is how are our academics going to do the research they

academics going to do the research they need to do to offer an unbi Fu so so

need to do to offer an unbi Fu so so those are the things I think about and

those are the things I think about and and I don't have great answers because

and I don't have great answers because it's not it's it's a complicated Nuance

it's not it's it's a complicated Nuance issue but what I do know is that uh you

issue but what I do know is that uh you can actually do a lot more harm by

can actually do a lot more harm by moving quickly and being tighter so just

moving quickly and being tighter so just I would say take your time and

I would say take your time and effectively let the market develop but

effectively let the market develop but also please add a dimension of being

also please add a dimension of being proest like I think it's got to be Pro

proest like I think it's got to be Pro competition Pro Innovation and proest

competition Pro Innovation and proest because there are a different set of

because there are a different set of rules that are happening in China and we

rules that are happening in China and we don't want to handicap our companies

don't want to handicap our companies here thank you vinky so George uh sorry

here thank you vinky so George uh sorry David we're going to come to this uh

David we're going to come to this uh topic that you brought up the open

topic that you brought up the open source and also uh the issue of

source and also uh the issue of Regulation specifically at a6c you've

Regulation specifically at a6c you've been very vocal about value of AI for

been very vocal about value of AI for the world uh but also skeptical about

the world uh but also skeptical about the role of Regulation specifically that

the role of Regulation specifically that it could reduce competition can you

it could reduce competition can you explain how at6 you were thinking about

explain how at6 you were thinking about that yeah absolutely um and it actually

that yeah absolutely um and it actually relates to some of the things that have

relates to some of the things that have already been said um because I I agree

already been said um because I I agree with a lot of what what Karen said um

with a lot of what what Karen said um not surprisingly I agree with a lot of

not surprisingly I agree with a lot of what vinky said too uh so to the point

what vinky said too uh so to the point on um you know taking a beat and a

on um you know taking a beat and a little bit of patience um specifically

little bit of patience um specifically in the foundation model side the market

in the foundation model side the market is so early that we actually have no

is so early that we actually have no idea how the value chain is going to

idea how the value chain is going to evolve like we don't actually know how

evolve like we don't actually know how many players in the foundation model

many players in the foundation model space are going to be relevant we don't

space are going to be relevant we don't know which ones they're going to be

know which ones they're going to be there's a major swing Factor if we allow

there's a major swing Factor if we allow open source to flourish versus if we

open source to flourish versus if we don't um you know moreover if you look

don't um you know moreover if you look at what the cloud companies are doing

at what the cloud companies are doing which to me they're they're synonymous

which to me they're they're synonymous with big Tech um from my vantage point

with big Tech um from my vantage point it seems to me like they are trying to

it seems to me like they are trying to commoditize that layer so they're

commoditize that layer so they're actually using their competitive power

actually using their competitive power to make sure that the economics are not

to make sure that the economics are not too attractive at the foundation model

too attractive at the foundation model side and if you look at what's happening

side and if you look at what's happening in the foundation model side sure

in the foundation model side sure Capital it has a lot of things that

Capital it has a lot of things that resemble big Tech um Capital requirement

resemble big Tech um Capital requirement lots of talent you know access to data

lots of talent you know access to data centers Etc um but prices are absolutely

centers Etc um but prices are absolutely plummeting like they are falling like a

plummeting like they are falling like a rock and that's great that's actually

rock and that's great that's actually very good for the market uh because it

very good for the market uh because it should get very valuable technology into

should get very valuable technology into more people's hands um the more that

more people's hands um the more that that technology ideally in open source

that technology ideally in open source format is available to people building

format is available to people building products

products the more likely consumers are going to

the more likely consumers are going to be able to benefit from them uh you know

be able to benefit from them uh you know at large um so I would also urge uh

at large um so I would also urge uh patients I would I would urge um you

patients I would I would urge um you know letting the foundation model

know letting the foundation model specifically Market evolve a little bit

specifically Market evolve a little bit for the time being regulate how people

for the time being regulate how people use those models as opposed to

use those models as opposed to regulating the models themselves JS you

regulating the models themselves JS you said something which I agree with fully

said something which I agree with fully they are very primitive so they think of

they are very primitive so they think of them as machines that do math equations

them as machines that do math equations um so coming up with a regulatory

um so coming up with a regulatory requirement for those in any form is

requirement for those in any form is more likely than not to produce an undue

more likely than not to produce an undue burden for smaller companies and produce

burden for smaller companies and produce you know some amount of work that bigger

you know some amount of work that bigger bigger tech companies are able to to

bigger tech companies are able to to deal with um so as far as

deal with um so as far as anti-competitive stuff that I'm seeing

anti-competitive stuff that I'm seeing in the market today goes um you're

in the market today goes um you're looking very closely at the cloud

looking very closely at the cloud companies uh for good reason um you know

companies uh for good reason um you know we know that these companies are very

we know that these companies are very ingrained in DC and the the political

ingrained in DC and the the political Western countries uh they have deep

Western countries uh they have deep lobbying efforts um but know that when

lobbying efforts um but know that when they come to you they they have their

they come to you they they have their objective in mind uh as well um

objective in mind uh as well um specifically on the open source piece if

specifically on the open source piece if I may just touch on why I actually think

I may just touch on why I actually think it's a good idea um uh VY I think you

it's a good idea um uh VY I think you were the one who said we have an open

were the one who said we have an open internet like that's fantastic it did it

internet like that's fantastic it did it actually didn't have to break that way

actually didn't have to break that way um you know you can see actually what

um you know you can see actually what the closed platforms look like and I

the closed platforms look like and I think many people in this room are very

think many people in this room are very unhappy with that outcome uh I know

unhappy with that outcome uh I know you're doing a lot of work to try and

you're doing a lot of work to try and combat it um but you know if you look at

combat it um but you know if you look at at regulations that are well- intended

at regulations that are well- intended like

like gdpr the biggest implication of gdpr is

gdpr the biggest implication of gdpr is actually it strengthened the larger

actually it strengthened the larger players because they were the only ones

players because they were the only ones able to deal with the the compliance

able to deal with the the compliance burden that came with it it so I would

burden that came with it it so I would urge uh a lot of caution in applying

urge uh a lot of caution in applying something similar uh at such an early

something similar uh at such an early stage at the foundation model level um

stage at the foundation model level um but again let's let's take the rules and

but again let's let's take the rules and regulations and the laws that we have in

regulations and the laws that we have in place and make sure that we are

place and make sure that we are regulating the usage of these

regulating the usage of these Technologies the

Technologies the applications um you know the threats

applications um you know the threats that that bad actors will be able to to

that that bad actors will be able to to to be better with them um you know on

to be better with them um you know on open source uh

open source uh specifically um you know one I think it

specifically um you know one I think it allows us to move faster um you know

allows us to move faster um you know look at the way Linux has developed it's

look at the way Linux has developed it's another one in addition to the open

another one in addition to the open internet um you know it's extraordinary

internet um you know it's extraordinary that we have those two outcomes that are

that we have those two outcomes that are open um so it allows us to move faster

open um so it allows us to move faster because there are more people working on

because there are more people working on it um there's greater

it um there's greater transparency um and it's not frankly

transparency um and it's not frankly controlled by three or four very

controlled by three or four very powerful companies that exert their

powerful companies that exert their monopolistic power um to use it to

monopolistic power um to use it to advance their their means um and uh and

advance their their means um and uh and you know potentially harm consumers so I

you know potentially harm consumers so I agree highest level I would urge uh

agree highest level I would urge uh caution I would I would urge letting the

caution I would I would urge letting the foundation model layer um solidify a

foundation model layer um solidify a little bit further let us understand who

little bit further let us understand who you know who has power who doesn't um

you know who has power who doesn't um but I would be looking very closely at

but I would be looking very closely at you know some of the the deals and

you know some of the the deals and things that um that the big Tech players

things that um that the big Tech players are doing can I just add one thing so

are doing can I just add one thing so David many things you covered I've been

David many things you covered I've been thinking a lot about what would be an

thinking a lot about what would be an analogous element we can use in this

analogous element we can use in this context I think when the federal

context I think when the federal government kind of bailed out the banks

government kind of bailed out the banks they forced all the large Banks they too

they forced all the large Banks they too big to fail Banks to create what's

big to fail Banks to create what's called a living Bill where basically

called a living Bill where basically they have to say what will happen if I

they have to say what will happen if I need to shut down in a safe way I think

need to shut down in a safe way I think that kind of self-reporting I think is a

that kind of self-reporting I think is a great framework I think we should force

great framework I think we should force all our foundational models and people

all our foundational models and people say what what what is your safety

say what what what is your safety procedures in place what are the things

procedures in place what are the things you're looking for we need that to be

you're looking for we need that to be and send to the documented and send it

and send to the documented and send it to the regulator so people have a sense

to the regulator so people have a sense because that's a good way of building

because that's a good way of building early warning signals and then forcing

early warning signals and then forcing people to build system I think it's very

people to build system I think it's very hard for Regulators to retroactively

hard for Regulators to retroactively come in some of these situations I think

come in some of these situations I think it's much better to proactively create a

it's much better to proactively create a reporting regime where say build your

reporting regime where say build your plant tell us exactly what you're going

plant tell us exactly what you're going to look for when will emergent

to look for when will emergent properties when emergent what will be an

properties when emergent what will be an emergent property if you're not familiar

emergent property if you're not familiar what that is is when a model starts

what that is is when a model starts showing behavior that we think is a sign

showing behavior that we think is a sign of intelligence Beyond it's called an

of intelligence Beyond it's called an emergent property what is your evidence

emergent property what is your evidence of emergent properties what is your plan

of emergent properties what is your plan what is your treasur safety plan

what is your treasur safety plan document that and commit to it because

document that and commit to it because that's a good way of essentially driving

that's a good way of essentially driving the free market and then over time

the free market and then over time Regulators can see if they need to add

Regulators can see if they need to add to that but that's a good start I mean I

to that but that's a good start I mean I I totally agree I one just quick caveat

I totally agree I one just quick caveat because I think when you say emergent

because I think when you say emergent properties and intell signs of

properties and intell signs of intelligence it like that's one of the

intelligence it like that's one of the things that I think stok some of the

things that I think stok some of the fear um these models are directed by us

fear um these models are directed by us they don't have an objective function

they don't have an objective function like we are the objective function so I

like we are the objective function so I think that's very important to

think that's very important to understand it's not like we're on the

understand it's not like we're on the precipice of some technological

precipice of some technological breakthrough where models are suddenly

breakthrough where models are suddenly going to have their own objective

going to have their own objective function that's not what's that's not

function that's not what's that's not the work that's going to be produced uh

the work that's going to be produced uh anytime in the near term so when people

anytime in the near term so when people talk about super intelligence and AGI

talk about super intelligence and AGI and things like that understand it for

and things like that understand it for what it is it's when a human has access

what it is it's when a human has access to the model with an objective function

to the model with an objective function they can accomplish more but it's

they can accomplish more but it's nothing beyond that and so I think that

nothing beyond that and so I think that should do a little bit to do away with

should do a little bit to do away with some of the the fears around safety uh

some of the the fears around safety uh that I think would act would maybe lead

that I think would act would maybe lead us to act a little more hastily than we

us to act a little more hastily than we otherwise would right thank you uh I

otherwise would right thank you uh I want to change uh while we focus is

want to change uh while we focus is still on the competition but I want to

still on the competition but I want to change the topic a little bit and uh

change the topic a little bit and uh similar questions from everyone but

similar questions from everyone but depending on uh like what your angle so

depending on uh like what your angle so I'm going to start with Karen you know

I'm going to start with Karen you know every few months a new model is released

every few months a new model is released and at least on the benchmarks it's a

and at least on the benchmarks it's a lot better and the whole technology is

lot better and the whole technology is rapidly evolving this morning we heard

rapidly evolving this morning we heard from Andrew and Percy about you know

from Andrew and Percy about you know like this agent uh workflow and Rag and

like this agent uh workflow and Rag and all these things that have such a

all these things that have such a tremendous impact overall the space is

tremendous impact overall the space is rapidly evolving so Karen uh it must be

rapidly evolving so Karen uh it must be difficult to oversee AI markets while

difficult to oversee AI markets while things are rapidly changing how does CMA

things are rapidly changing how does CMA ensure it has the capability to

ensure it has the capability to understand the complex emerging Tech to

understand the complex emerging Tech to keep up and tell us a bit about your

keep up and tell us a bit about your program around

program around that yeah thank you and I uh you know I

that yeah thank you and I uh you know I think this is we I think it's important

think this is we I think it's important not to be complacent in this space I

not to be complacent in this space I mean this is a genuinely um complex uh

mean this is a genuinely um complex uh ecosystem the technology itself is

ecosystem the technology itself is complex um the the way that this Market

complex um the the way that this Market is working you know has a complexity to

is working you know has a complexity to it um and you know as David you've noted

it um and you know as David you've noted already the the value chain is still

already the the value chain is still evolving and you know we're continuing

evolving and you know we're continuing to see as I say that drum beat of

to see as I say that drum beat of disruption of of innovation and activity

disruption of of innovation and activity and so this is not a space for

and so this is not a space for complacency um I think it is really

complacency um I think it is really critical that uh we as an agency I'll

critical that uh we as an agency I'll speak for the CMA we as an agency um

speak for the CMA we as an agency um absolutely are um in this space with

absolutely are um in this space with some in-house technical capability an

some in-house technical capability an ability to really understand uh these

ability to really understand uh these models sufficiently these Technologies

models sufficiently these Technologies sufficiently and this is true generally

sufficiently and this is true generally for technology across markets uh and an

for technology across markets uh and an ability and I think this is really

ability and I think this is really critical you know you need that you need

critical you know you need that you need that technical lens that understanding

that technical lens that understanding of the technology But ultimately we need

of the technology But ultimately we need to get to implications for consumers and

to get to implications for consumers and competition which is our mandate in the

competition which is our mandate in the UK and so uh you know I'm my uh first

UK and so uh you know I'm my uh first degree and my PhD were in economics

degree and my PhD were in economics economics actually and understanding the

economics actually and understanding the economics of the market and the way that

economics of the market and the way that um technological change shifts the

um technological change shifts the economics of markets is important for

economics of markets is important for any uh Technology based disruption um

any uh Technology based disruption um you obviously need um to understand uh

you obviously need um to understand uh the law and have legal analysis you need

the law and have legal analysis you need need strategic business and financial

need strategic business and financial analysis to really understand not just

analysis to really understand not just the technology itself but how that will

the technology itself but how that will form part of a strategy brought to

form part of a strategy brought to Market what is the business model how

Market what is the business model how will firms whether it's little Tech or

will firms whether it's little Tech or big Tech be looking to make uh make

big Tech be looking to make uh make their money um and get a return on these

their money um and get a return on these kinds of Investments what will be the

kinds of Investments what will be the considerations in key verticals and

considerations in key verticals and areas for for fine-tuning and deployment

areas for for fine-tuning and deployment of these models uh and so on and so

of these models uh and so on and so forth

forth and there are many other skill sets

and there are many other skill sets beside but there's a there's a an

beside but there's a there's a an important space I think for taking

important space I think for taking what's going you know scanning the

what's going you know scanning the markets in that prioritized way um

markets in that prioritized way um bringing that into an agency and doing

bringing that into an agency and doing the sense making and the problem solving

the sense making and the problem solving around that and really working through

around that and really working through into the implications and it's it's when

into the implications and it's it's when you have those implications and the

you have those implications and the evidence and the analysis around that

evidence and the analysis around that that you can really shape evidence based

that you can really shape evidence based decision making right through to any

decision making right through to any actions you would take uh in the market

actions you would take uh in the market and so actually you know asked about my

and so actually you know asked about my program or our program at the CMA I

program or our program at the CMA I think I see uh what we're doing as

think I see uh what we're doing as important but part of a bigger hole

important but part of a bigger hole inside the agency we need to be

inside the agency we need to be collaborating and we collaborate you

collaborating and we collaborate you know we're looking to embed this as an

know we're looking to embed this as an additional capability very deeply across

additional capability very deeply across the life cycle of the cma's work from

the life cycle of the cma's work from upfront Horizon scanning right through

upfront Horizon scanning right through to sort of understanding the impact of

to sort of understanding the impact of some of our interventions um and shaping

some of our interventions um and shaping new approaches we need to be um

new approaches we need to be um collaborating a lot internally and doing

collaborating a lot internally and doing that mix of reactive and proactive work

that mix of reactive and proactive work working on all the major digital cases

working on all the major digital cases using data and understanding algorithmic

using data and understanding algorithmic systems but also driving proactive

systems but also driving proactive thematic work on issues like Ai and

thematic work on issues like Ai and Entre and uh online Choice architecture

Entre and uh online Choice architecture I mentioned but actually you know we

I mentioned but actually you know we don't operate in a vacuum so then when

don't operate in a vacuum so then when you step back as as an agency like

you step back as as an agency like ourselves um there are intersections

ourselves um there are intersections with other important policy OB

with other important policy OB objectives and data protection was

objectives and data protection was touched on a moment ago we're not the

touched on a moment ago we're not the data um protection uh agency in the UK

data um protection uh agency in the UK we are part of something in the UK

we are part of something in the UK called the digital regulation

called the digital regulation cooperation Forum or drcf as we call it

cooperation Forum or drcf as we call it um that was something that we formed a

um that was something that we formed a couple of years ago together with the uh

couple of years ago together with the uh Ico the data protection regulator ofcom

Ico the data protection regulator ofcom the media regulator and also the FCA

the media regulator and also the FCA which is the financial conduct Authority

which is the financial conduct Authority and I worked previously at the financial

and I worked previously at the financial conduct Authority so major agencies with

conduct Authority so major agencies with a major involvement in the digital

a major involvement in the digital markets and we come together in that

markets and we come together in that space and we are able to share knowledge

space and we are able to share knowledge and ensure that the regulatory landscape

and ensure that the regulatory landscape overall in the UK is coherent and will

overall in the UK is coherent and will promote Innovation and promote effective

promote Innovation and promote effective competition and allow innovators to come

competition and allow innovators to come into the market and really navigate uh

into the market and really navigate uh that landscape so recently as just one

that landscape so recently as just one example we have set up a pilot AI um Hub

example we have set up a pilot AI um Hub as the drcf um and we have innovators

as the drcf um and we have innovators starting already to come through that

starting already to come through that Hub and get support um and you know sort

Hub and get support um and you know sort of direct engagement on how they can

of direct engagement on how they can understand how to bring a new model to

understand how to bring a new model to Market and some of the considerations

Market and some of the considerations across those different mandates and then

across those different mandates and then I would just end by saying that um of

I would just end by saying that um of course these are Global firms often that

course these are Global firms often that we're talking about global issues and

we're talking about global issues and consideration so it's very important

consideration so it's very important that we are um engaging actively

that we are um engaging actively internationally so events like uh today

internationally so events like uh today here at Stanford are incredibly valuable

here at Stanford are incredibly valuable and we engage a lot across the ecosystem

and we engage a lot across the ecosystem globally um we have something called the

globally um we have something called the international competition Network um

international competition Network um that brings together the competition

that brings together the competition agencies across the world that network

agencies across the world that network uh there we recently have established a

uh there we recently have established a technologist Forum which is an excellent

technologist Forum which is an excellent forum for allowing us a space um as

forum for allowing us a space um as agencies to um share insights about how

agencies to um share insights about how you build capability and capacity in

you build capability and capacity in this space as well as discussing some of

this space as well as discussing some of what we're seeing in markets and um you

what we're seeing in markets and um you know the um the considerations there and

know the um the considerations there and I would just say there was a mention of

I would just say there was a mention of academics um you know this isn't my

academics um you know this isn't my first visit here to Stanford there's

first visit here to Stanford there's obviously excellent academic expertise

obviously excellent academic expertise here at Stanford in many other

here at Stanford in many other universities too we engage a lot with

universities too we engage a lot with academic experts with our foundation

academic experts with our foundation models work that's been a um a a a core

models work that's been a um a a a core channel that we've had open uh

channel that we've had open uh throughout that work and continue to

throughout that work and continue to benefit from thank you Karen so I see

benefit from thank you Karen so I see that we have only five minutes um I I'm

that we have only five minutes um I I'm going to actually open for questions

going to actually open for questions from audience and then come back yes

from audience and then come back yes have

have one hi um I have a question related to

one hi um I have a question related to the resources that we talk about and we

the resources that we talk about and we didn't uh I I'm not an Anti-Trust expert

didn't uh I I'm not an Anti-Trust expert but we are talking about a resource

but we are talking about a resource constraint issue and as we see these

constraint issue and as we see these models grow uh much more Out Of Reach

models grow uh much more Out Of Reach particularly for Academia but also in

particularly for Academia but also in Industry itself we seem to be really

Industry itself we seem to be really focused on competition aspects but we're

focused on competition aspects but we're talking about either Talent or data or

talking about either Talent or data or chips but we we're not really talking

chips but we we're not really talking about is the energy constraints related

about is the energy constraints related to this and to build Frontier Foundation

to this and to build Frontier Foundation models what will be the com competition

models what will be the com competition environment look like as the energy gets

environment look like as the energy gets much more constraining on these models

much more constraining on these models we see chat GPT one of the latest

we see chat GPT one of the latest versions built I believe outside of De

versions built I believe outside of De Moine pulling on the grids what will

Moine pulling on the grids what will happen in this particular environment

happen in this particular environment and does that make it much more

and does that make it much more competitive for only a few select

competitive for only a few select players to be in the

players to be in the space thank you wants to take this um

space thank you wants to take this um I'm happy to take I mean I'm not an

I'm happy to take I mean I'm not an energy expert but clearly these models

energy expert but clearly these models are at the scale in which they

are at the scale in which they operating if you were to draw the line

operating if you were to draw the line You' say Well they're going to use too

You' say Well they're going to use too much energy and we don't have enough

much energy and we don't have enough energy to fund that but the thing that

energy to fund that but the thing that I've known about technology is that you

I've known about technology is that you can't draw that line because it's always

can't draw that line because it's always going to be Innovation so there's a

going to be Innovation so there's a whole bunch of people working on how to

whole bunch of people working on how to make these models more energy less

make these models more energy less energy intensive it's just hard to

energy intensive it's just hard to predict the scope to it but you

predict the scope to it but you absolutely right in pointing out that

absolutely right in pointing out that these are all elements in which it's

these are all elements in which it's going to become only a small there's

going to become only a small there's really a huge capital intensivity in

really a huge capital intensivity in terms of both how much Capital you need

terms of both how much Capital you need to train these models then there access

to train these models then there access to that GPU and that's on shortage and

to that GPU and that's on shortage and then there's how much energy you're

then there's how much energy you're going to take so all of these Dimensions

going to take so all of these Dimensions point to a world in which they're going

point to a world in which they're going to be less number of models rather than

to be less number of models rather than more um I can add to that as well I

more um I can add to that as well I think what if you look at the world of

think what if you look at the world of foundation models um and just you know

foundation models um and just you know dial Back Time 6 months ago a lot of the

dial Back Time 6 months ago a lot of the capabilities that we were using were per

capabilities that we were using were per capita much more expensive than they are

capita much more expensive than they are right now from an energy consumption

right now from an energy consumption perspective

perspective and if you look at Foundation models

and if you look at Foundation models they're sort of splitting into smaller

they're sort of splitting into smaller models and really large models if you if

models and really large models if you if you're able to break down problems into

you're able to break down problems into task specific components those are are a

task specific components those are are a lot more energy efficient have we gotten

lot more energy efficient have we gotten to the point where we're able to merge

to the point where we're able to merge these two together in a great way where

these two together in a great way where you're able to use reasoning and

you're able to use reasoning and planning which is going to be more

planning which is going to be more expensive to versus task specific model

expensive to versus task specific model separately not yet but there are I think

separately not yet but there are I think good signs that um there is efficiencies

good signs that um there is efficiencies being out at the task level this is this

being out at the task level this is this is not at all what this panel is about

is not at all what this panel is about but um nuclear would be a great solution

but um nuclear would be a great solution to be able to catch up um and so you

to be able to catch up um and so you know we're investors self-interested

know we're investors self-interested we're investors in a nuclear company but

we're investors in a nuclear company but uh but I do happen to think that that is

uh but I do happen to think that that is the future of being able to keep up with

the future of being able to keep up with the energy requirements of this new

the energy requirements of this new technology maybe just time for one more

technology maybe just time for one more question I see one hand over

there thank you Adam star I'm an MBA at Stanford uh I have a question for the

Stanford uh I have a question for the venture capitalists we Karen talked

venture capitalists we Karen talked about some of the way areas where there

about some of the way areas where there are risks of anti-competitive pressure

are risks of anti-competitive pressure are there areas that you are seeing a

are there areas that you are seeing a lack of investment due to monopolistic

lack of investment due to monopolistic Pressure David do you want

to look things that have so I think the large language models is

so I think the large language models is the thing obviously to look at um this

the thing obviously to look at um this is why I'm So Pro open source is because

is why I'm So Pro open source is because I think it is a way for other people who

I think it is a way for other people who don't have as much resources to be able

don't have as much resources to be able to build on top of something that

to build on top of something that someone with a lot of resources are able

someone with a lot of resources are able to build um the capital requirements of

to build um the capital requirements of building large language models are going

building large language models are going to be such that there's not going to be

to be such that there's not going to be you know tons and tons of players just

you know tons and tons of players just from a market structure standpoint um I

from a market structure standpoint um I I think the potential for future Bad

I think the potential for future Bad actors in the creation of New Market

actors in the creation of New Market structures is is the big unknown

structures is is the big unknown um you know we

um you know we are very excited about the application

are very excited about the application layer so if you think about like

layer so if you think about like Foundation models um there's going to be

Foundation models um there's going to be only so few players I mentioned I don't

only so few players I mentioned I don't actually know what the value capture is

actually know what the value capture is going to be there actually be might be a

going to be there actually be might be a lot of value capture or no value capture

lot of value capture or no value capture at that layer I am extremely confident

at that layer I am extremely confident that at the application layer there's

that at the application layer there's going to be a tremendous amount of value

going to be a tremendous amount of value to be captured and those are the kinds

to be captured and those are the kinds of things that we would be excited about

of things that we would be excited about investing in but we are so early in this

investing in but we are so early in this curve um that we're just seeing only the

curve um that we're just seeing only the initial signs of what people are

initial signs of what people are actually going to build so like you you

actually going to build so like you you know AI chatbots for example I don't

know AI chatbots for example I don't think the future of this technology in

think the future of this technology in its fullest form is in Jackpot form I

its fullest form is in Jackpot form I think it's going to be a reimagination

think it's going to be a reimagination of how we do work um and you know what

of how we do work um and you know what kind of productivity productivity gains

kind of productivity productivity gains we can get as consumers but all that is

we can get as consumers but all that is very very very

very very very early so the short answer from from my

early so the short answer from from my point is no not yet I I think the most

point is no not yet I I think the most important thing is we need to have lot

important thing is we need to have lot of capital flows because look in essence

of capital flows because look in essence you want if you look at every sort of

you want if you look at every sort of Technology cycle you want people to find

Technology cycle you want people to find a lot of companies now here's a fact 95%

a lot of companies now here's a fact 95% of all the AI companies we funding are

of all the AI companies we funding are going to go down to zero now 5% are

going to go down to zero now 5% are going to turn out to be incredible

going to turn out to be incredible Investments I my wife always says well

Investments I my wife always says well just invested the 5% promise I don't

just invested the 5% promise I don't know which is the 5% so easy and and and

know which is the 5% so easy and and and and but it is for us beneficial as a

and but it is for us beneficial as a society to create an environment in

society to create an environment in which we can have Capital flows come in

which we can have Capital flows come in there that's the benefit of the

there that's the benefit of the capitalistic system and so I would just

capitalistic system and so I would just say as long as the capital flows come in

say as long as the capital flows come in we'll self sort this

we'll self sort this out um so you know we are over time I

out um so you know we are over time I know there are more questions uh we wish

know there are more questions uh we wish we had more time to hear from our

we had more time to hear from our experts uh but hope you found the

experts uh but hope you found the discussion informative and

discussion informative and thought-provoking let's give a big thank

thought-provoking let's give a big thank to our

panelists thank you for joining us and enjoy the rest of the workshop

all right thank you so much professor bayti and all the panel speakers um next

bayti and all the panel speakers um next we have our afternoon keynote I know

we have our afternoon keynote I know many of you are really looking forward

many of you are really looking forward to hearing from our next speaker Dr

to hearing from our next speaker Dr condalisa Rice uh but before we properly

condalisa Rice uh but before we properly introduce her uh I'd like to introduce

introduce her uh I'd like to introduce the moderator uh my super colleague

the moderator uh my super colleague Professor Greg Roston Professor Roston

Professor Greg Roston Professor Roston is the Gordon K senior P at Seer and

is the Gordon K senior P at Seer and director of the public policy program at

director of the public policy program at Stanford uh he teaches courses on

Stanford uh he teaches courses on competition policy and strategy

competition policy and strategy intellectual property and personal

intellectual property and personal finance so please join me in welcoming

finance so please join me in welcoming Greg to the

stage thanks thank you I'm uh thrilled to have

thanks thank you I'm uh thrilled to have the honor to introduce secretary condar

the honor to introduce secretary condar rice uh my former John chovin always

rice uh my former John chovin always said the the better known someone is the

said the the better known someone is the shorter the introduction should be I'm

shorter the introduction should be I'm not going to stop right there but what

not going to stop right there but what I'm going to do is let you read the bio

I'm going to do is let you read the bio and you all know everything about her

and you all know everything about her except for something that my view as the

except for something that my view as the I'm not only at Seer I'm also the

I'm not only at Seer I'm also the director of the public policy program

director of the public policy program and you know there are lots of places to

and you know there are lots of places to do academic research and there's lots of

do academic research and there's lots of places to do policy stuff but the big

places to do policy stuff but the big thing at Stanford is to be involved with

thing at Stanford is to be involved with students and I've had the honor of

students and I've had the honor of having secretary rice come and talk to

having secretary rice come and talk to my classes

my classes I know students who had her as an

I know students who had her as an advisor and she's amazing this is you

advisor and she's amazing this is you know you might think of an University is

know you might think of an University is being left leaning and not open to ideas

being left leaning and not open to ideas but she has an incredible waiting list

but she has an incredible waiting list for her classes high demand as an

for her classes high demand as an advisor and she's fantastic in a

advisor and she's fantastic in a classroom and you're going to get to

classroom and you're going to get to enjoy that for the next 30 minutes so

enjoy that for the next 30 minutes so welcome Dr Rice with me

I'm trying to figure out where the middle middle these I'm trying to figure

middle middle these I'm trying to figure out where the middle

was uh so thanks a lot for for coming I know um you know today we've had this uh

know um you know today we've had this uh conversation about

conversation about competition and as an antitrust guy I

competition and as an antitrust guy I think of competition as competition

think of competition as competition between firms and I know in your most of

between firms and I know in your most of your life competition has not

your life competition has not necessarily been Market competition

necessarily been Market competition among firms but competition among count

among firms but competition among count and we heard uh vanki ganess in the last

and we heard uh vanki ganess in the last session talk about having pro-innovation

session talk about having pro-innovation Pro competition and pro us I think is

Pro competition and pro us I think is what you said and thinking about how do

what you said and thinking about how do you think about how AI is going to

you think about how AI is going to affect competition yes well first of all

affect competition yes well first of all thanks very much for having me and I

thanks very much for having me and I know it's been a really interesting

know it's been a really interesting discussion um I I think it's awfully

discussion um I I think it's awfully important that we have these for because

important that we have these for because I think a lot of people in the policy

I think a lot of people in the policy world uh have learned to spell AI but

world uh have learned to spell AI but there still not perhaps certain uh about

there still not perhaps certain uh about its implications and so continuing to

its implications and so continuing to talk about it is is really important

talk about it is is really important you're right Greg that competition for

you're right Greg that competition for me uh generally throughout my career has

me uh generally throughout my career has meant something else and I'll get to

meant something else and I'll get to that in a moment but one of the reasons

that in a moment but one of the reasons that um I believe that uh the United

that um I believe that uh the United States uh with a proper with proper

States uh with a proper with proper attention to its Innovation uh ecosystem

attention to its Innovation uh ecosystem uh which has an academic element it has

uh which has an academic element it has a private sector element and of course

a private sector element and of course then has uh government Poli policy uh

then has uh government Poli policy uh that attention to that will U give the

that attention to that will U give the United States the advantage uh in the

United States the advantage uh in the competition that I'm about to describe

competition that I'm about to describe is that uh it is in fact the competition

is that uh it is in fact the competition the competition of ideas the competition

the competition of ideas the competition of capital the competition for markets

of capital the competition for markets that has driven our economy and I think

that has driven our economy and I think made us the most powerful economy in the

made us the most powerful economy in the world and I think this uh these Frontier

world and I think this uh these Frontier Technologies it will be no different uh

Technologies it will be no different uh the very fact that you will have several

the very fact that you will have several uh entities

uh entities competing for uh Supremacy uh rather

competing for uh Supremacy uh rather than a single dictat from uh the

than a single dictat from uh the government for it I think gives us a

government for it I think gives us a tremendous advantage and so I hope that

tremendous advantage and so I hope that nothing that we do ab Bridges our

nothing that we do ab Bridges our ability to let uh competition Drive

ability to let uh competition Drive innovation in the way that it has in the

innovation in the way that it has in the United States for many many years um of

United States for many many years um of course the other big competitor is uh

course the other big competitor is uh China and um I would put it this way

China and um I would put it this way that we're really actually talking about

that we're really actually talking about an arms race now in technology and arms

an arms race now in technology and arms race in Frontier Technologies we're

race in Frontier Technologies we're talking about AI today but there's an

talking about AI today but there's an arms race in synthetic biology there's

arms race in synthetic biology there's an arms race in robotics there's an arms

an arms race in robotics there's an arms race in space um and it is really

race in space um and it is really between uh two countries that have

between uh two countries that have unfortunately over recent years become

unfortunately over recent years become more adversarial in uh their

more adversarial in uh their relationship and so thinking of it as

relationship and so thinking of it as somebody uh is going to try to win I is

somebody uh is going to try to win I is not really out of line even though uh

not really out of line even though uh maybe 20 years or so ago when we had a

maybe 20 years or so ago when we had a kind of integrationist narrative about

kind of integrationist narrative about China in the international economy we

China in the international economy we might not have thought about it that way

might not have thought about it that way um I I've said to many of my Chinese

um I I've said to many of my Chinese friends and I think I'm still considered

friends and I think I'm still considered somebody who um wants to see a better

somebody who um wants to see a better relationship with China we had a very

relationship with China we had a very good relationship with China in the Bush

good relationship with China in the Bush Administration but I've said that

Administration but I've said that perhaps one of the let me just put it

perhaps one of the let me just put it dumbest speeches I think I've ever heard

dumbest speeches I think I've ever heard given by a leader was the one which she

given by a leader was the one which she ping uh said that China was going to

ping uh said that China was going to surpass the United States in Frontier

surpass the United States in Frontier Technologies like Ai and uh and Quantum

Technologies like Ai and uh and Quantum Computing because what did he think was

Computing because what did he think was going to happen uh it got our backs up

going to happen uh it got our backs up uh we said oh that's what this is all

uh we said oh that's what this is all about and then the Chinese came up with

about and then the Chinese came up with this thing called civil military Fusion

this thing called civil military Fusion so we then said oh so let me get this

so we then said oh so let me get this right the Chinese Innovation which we

right the Chinese Innovation which we are helping to fuel through investment

are helping to fuel through investment in Chinese uh startups and Chinese

in Chinese uh startups and Chinese companies is going to be handed over to

companies is going to be handed over to the pla so that they can uh expel us

the pla so that they can uh expel us from the Indo Pacific well that doesn't

from the Indo Pacific well that doesn't sound like a very good idea and So what

sound like a very good idea and So what had the potential I think to uh build on

had the potential I think to uh build on the open uh EOS ecosystem the open

the open uh EOS ecosystem the open nature of innovation worldwide uh has

nature of innovation worldwide uh has become very hardened into two camps and

become very hardened into two camps and I think we are actually decoupling from

I think we are actually decoupling from China quite rapidly and quite

China quite rapidly and quite dramatically and that uh puts this

dramatically and that uh puts this competition uh in a different a

competition uh in a different a different framework so do you see uh the

different framework so do you see uh the government policies of sort of

government policies of sort of restricting trade and you know Frontier

restricting trade and you know Frontier technology with China to be effective in

technology with China to be effective in changing the way things are going to

changing the way things are going to work or is this just delaying the

work or is this just delaying the inevitable of China taking over and

inevitable of China taking over and developing its own things uh how do you

developing its own things uh how do you see the difference in what we're doing

see the difference in what we're doing as a government policy well let me

as a government policy well let me separate it into three areas one of

separate it into three areas one of which um I'm concerned that U maybe the

which um I'm concerned that U maybe the government should try and stay out of if

government should try and stay out of if at all possible but let look at the

at all possible but let look at the front here um I'm not surprised that um

front here um I'm not surprised that um that sanctions and uh restrictions uh

that sanctions and uh restrictions uh trade restrictions even restrictions on

trade restrictions even restrictions on inbound and outbound investment have

inbound and outbound investment have become part of the toolkit for deal in

become part of the toolkit for deal in with uh what is now viewed as China's

with uh what is now viewed as China's challenge to American technological

challenge to American technological Supremacy and therefore to American uh

Supremacy and therefore to American uh leadership because after all uh

leadership because after all uh technology is probably going to be the

technology is probably going to be the dominant factor in what country uh is

dominant factor in what country uh is most uh is is most powerful is most

most uh is is most powerful is most dominant in the International System uh

dominant in the International System uh this is a very different challenge you

this is a very different challenge you know people often talk about well it's

know people often talk about well it's cold war too but of course the Soviet

cold war too but of course the Soviet Union was a military giant but it was a

Union was a military giant but it was a technological and economic and so

technological and economic and so when you talk about China and you look

when you talk about China and you look that they have all three of those

that they have all three of those elements military power technological

elements military power technological power economic power it's not surprising

power economic power it's not surprising that the United States is going to try

that the United States is going to try to restrict uh access to uh American

to restrict uh access to uh American technologies that can fuel China's uh

technologies that can fuel China's uh progress uh it will have I think

progress uh it will have I think temporary we don't know how long but

temporary we don't know how long but temporary effect for instance I'm told

temporary effect for instance I'm told by some people who are very high up in

by some people who are very high up in in Chinese Tech uh that generative AI

in Chinese Tech uh that generative AI just not not possible uh you know if we

just not not possible uh you know if we deny the Nvidia chip uh it's very

deny the Nvidia chip uh it's very difficult to do the kinds of things that

difficult to do the kinds of things that you do at the high end so for some time

you do at the high end so for some time yes but China is a very is a very uh

yes but China is a very is a very uh Innovative creative smart population it

Innovative creative smart population it has it's putting a lot of resources into

has it's putting a lot of resources into this from the top and uh this um lead if

this from the top and uh this um lead if you will is not going to last uh forever

you will is not going to last uh forever I can't tell you how long it's going to

I can't tell you how long it's going to last but it's not going to last forever

last but it's not going to last forever um there was a couple of years ago that

um there was a couple of years ago that we were talking about the fact that they

we were talking about the fact that they could train so much data because they

could train so much data because they didn't have privacy concerns and so

didn't have privacy concerns and so forth was going to give them the lead

forth was going to give them the lead nobody saw generative coming now if in

nobody saw generative coming now if in fact I'm right and the lead is uh is

fact I'm right and the lead is uh is temporary then uh what we have to do is

temporary then uh what we have to do is to try to continue to in out innovate is

to try to continue to in out innovate is to be in Leaps and Bounds going ahead

to be in Leaps and Bounds going ahead which is why my second point about

which is why my second point about government policy uh even if you're

government policy uh even if you're trying to delay China's uh China's uh

trying to delay China's uh China's uh progress uh the second part is Do no

progress uh the second part is Do no harm to our own Innovation System and

harm to our own Innovation System and that's where I worry because Regulators

that's where I worry because Regulators will regulate even if they don't know

will regulate even if they don't know what they're regulating and one of the

what they're regulating and one of the things that we're trying to do U at

things that we're trying to do U at Hoover is to that with Jennifer whittam

Hoover is to that with Jennifer whittam the dean of engineering we've created

the dean of engineering we've created something called the emerging Stanford

something called the emerging Stanford emerging technology review just to try

emerging technology review just to try to help policy makers understand what

to help policy makers understand what the frontiers of Technology look like to

the frontiers of Technology look like to try to demystify some of it so that fear

try to demystify some of it so that fear doesn't drive regulation I heard

doesn't drive regulation I heard somebody say recently well we have to be

somebody say recently well we have to be sure that we stop anything that's

sure that we stop anything that's harmful and I said how are you going to

harmful and I said how are you going to know what's harmful you're almost always

know what's harmful you're almost always going to mispredict what Innovation

going to mispredict what Innovation might look like who would have thought

might look like who would have thought that one of the biggest problems with

that one of the biggest problems with social media platform would be election

social media platform would be election interference so we can't second guess

interference so we can't second guess what that will look like and so I worry

what that will look like and so I worry that overregulation

that overregulation um particular some some of that I that I

um particular some some of that I that I see from Europe uh will actually stall

see from Europe uh will actually stall our Innovation and not allow us to

our Innovation and not allow us to outrace and then the third part of

outrace and then the third part of government policy that I hope we never

government policy that I hope we never see is um there was few years ago and

see is um there was few years ago and it's still there echo of it interference

it's still there echo of it interference in what we do in

in what we do in universities um I I'm not going to have

universities um I I'm not going to have nationality tests in my classrooms and

nationality tests in my classrooms and we shouldn't have nationality tests in

we shouldn't have nationality tests in our Labs um I've said to uh people in

our Labs um I've said to uh people in our intelligence agencies I understand

our intelligence agencies I understand the implications of Ip theft I

the implications of Ip theft I understand the implications of um the

understand the implications of um the Civil military Fusion programs of China

Civil military Fusion programs of China but uh if somebody you believe is really

but uh if somebody you believe is really working for the pla uh don't give them a

working for the pla uh don't give them a Visa and we won't admit them but don't

Visa and we won't admit them but don't try to turn universities into

try to turn universities into intelligence agents genes and so I

intelligence agents genes and so I separate government policy into yes we

separate government policy into yes we should try to slow China's progress to

should try to slow China's progress to the degree we can we should recognize

the degree we can we should recognize that that's temporary so secondarily

that that's temporary so secondarily let's not uh con let's not uh constrain

let's not uh con let's not uh constrain our own Innovation and third uh let's

our own Innovation and third uh let's try to keep open what has really worked

try to keep open what has really worked for us which is the um academic um

for us which is the um academic um ecosystem so I'm not sure which

ecosystem so I'm not sure which direction to go because you've brought

direction to go because you've brought up so many great issues uh let me let me

up so many great issues uh let me let me start with with the just the EU we had

start with with the just the EU we had the uh vice president of the European

the uh vice president of the European commission spoke this morning and talked

commission spoke this morning and talked about what regulations they're doing and

about what regulations they're doing and you expressed some concern with what the

you expressed some concern with what the EU is doing and I'd love to hear your

EU is doing and I'd love to hear your concern about what they're doing and

concern about what they're doing and what we can learn from that and what

what we can learn from that and what we're worried about well it's still

we're worried about well it's still evolving but um I've heard uh about

evolving but um I've heard uh about trying to judge for instance how to stop

trying to judge for instance how to stop harmful um I well I just don't even know

harmful um I well I just don't even know how to begin to think about that and uh

how to begin to think about that and uh it concerns me that the rush to regulate

it concerns me that the rush to regulate uh you know we might want to live in

uh you know we might want to live in this ecosystem for a little while uh and

this ecosystem for a little while uh and things will emerge that perhaps do need

things will emerge that perhaps do need regulation but if you're trying to

regulation but if you're trying to prefigure regulation for something that

prefigure regulation for something that is evolving this quickly you're almost

is evolving this quickly you're almost always going to make mistakes and so

always going to make mistakes and so I've been concerned about the rush to

I've been concerned about the rush to regulate I'm also hopeful that the

regulate I'm also hopeful that the European uh Union and the European

European uh Union and the European commission will understand that a lot of

commission will understand that a lot of the Innovation is actually not in

the Innovation is actually not in Europe um and so you could have a

Europe um and so you could have a separation from the regulatory

separation from the regulatory structures and where the Innovation is

structures and where the Innovation is actually taking place this is where the

actually taking place this is where the absence of Great Britain in Europe is a

absence of Great Britain in Europe is a big problem because the Innovation is

big problem because the Innovation is taking place in in uh the UK in places

taking place in in uh the UK in places like Oxford Deep Mind so forth and so

like Oxford Deep Mind so forth and so how to penetrate the conversation in the

how to penetrate the conversation in the EC and the EU about regulation with

EC and the EU about regulation with people who are actually part of the

people who are actually part of the Innovation uh Sy Innovation ecosystem I

Innovation uh Sy Innovation ecosystem I think is a real challenge for us um I've

think is a real challenge for us um I've been asked all the time about uh are

been asked all the time about uh are there some kind of international

there some kind of international Frameworks that we could use and I I

Frameworks that we could use and I I often say let let's start with

often say let let's start with like-minded countries uh let's start

like-minded countries uh let's start with countries that share values and

with countries that share values and obviously we share values with the

obviously we share values with the members of the European Union uh with

members of the European Union uh with the UK I would add that there are

the UK I would add that there are technologically sophisticated countries

technologically sophisticated countries that uh maybe not yet in AI but have

that uh maybe not yet in AI but have certainly the kind of engineering

certainly the kind of engineering capability in smarts India is a country

capability in smarts India is a country that ought to be a part of this

that ought to be a part of this discussion um you want to talk about

discussion um you want to talk about scaling uh a technology a biometric ID

scaling uh a technology a biometric ID for 1.3 billion people that's attached

for 1.3 billion people that's attached to a bank account I was in New Delhi

to a bank account I was in New Delhi recently and um I was at a conference

recently and um I was at a conference and they were describing how they use

and they were describing how they use the biometric ID to uh manage Co

the biometric ID to uh manage Co response and uh I said well that's

response and uh I said well that's really amazing you know if Tom Stevenson

really amazing you know if Tom Stevenson goes for his shot and they say no Tom

goes for his shot and they say no Tom your biometric ID says it should be two

your biometric ID says it should be two weeks from now and so I said yeah but

weeks from now and so I said yeah but here's how a really sophisticated

here's how a really sophisticated country handles this and I held up my

country handles this and I held up my tattered little vaccine card

tattered little vaccine card of which I've had two or three because I

of which I've had two or three because I keep losing them and so there are other

keep losing them and so there are other countries that can be Japan there are

countries that can be Japan there are countries that can be a part of this

countries that can be a part of this discussion that are like-minded I would

discussion that are like-minded I would start there so how so I I I was at a

start there so how so I I I was at a conference last week where you spoke

conference last week where you spoke about you know when you were Secretary

about you know when you were Secretary of State and you had these uh

of State and you had these uh telecommunications conferences and

telecommunications conferences and you've looked around to see who wasn't

you've looked around to see who wasn't doing anything and sent them off to

doing anything and sent them off to these negotiations kill me as a telec

these negotiations kill me as a telec person that but then I went to my

person that but then I went to my computer yesterday and saw there was

computer yesterday and saw there was another working group for a telecom

another working group for a telecom International Telecom conference and my

International Telecom conference and my eyes glazed over how important are these

eyes glazed over how important are these sort of you know deep under the details

sort of you know deep under the details uh organizations for thinking about how

uh organizations for thinking about how you would have international cooperation

you would have international cooperation and trying to get this idea of you know

and trying to get this idea of you know do you get India do you get the EU do

do you get India do you get the EU do you get everybody or do you start with a

you get everybody or do you start with a small number and how how did these

small number and how how did these organizations work to help you get there

organizations work to help you get there yeah I repeat I start with a small

yeah I repeat I start with a small number but to your point Greg so it is

number but to your point Greg so it is absolutely true you know when the ITC or

absolutely true you know when the ITC or somebody was about to have a a

somebody was about to have a a conference um I would say I really can't

conference um I would say I really can't afford to send somebody who's working on

afford to send somebody who's working on something important let me find somebody

something important let me find somebody who can go um and my guess is that that

who can go um and my guess is that that was happening among my colleagues in a

was happening among my colleagues in a number of countries but the one country

number of countries but the one country that that wasn't happening in was China

that that wasn't happening in was China and so they were sending people to these

and so they were sending people to these conferences particularly un conferences

conferences particularly un conferences where a lot of the rules rules making

where a lot of the rules rules making takes place and so the rules making gets

takes place and so the rules making gets embedded by people who sit there and go

embedded by people who sit there and go through you know the 48 hour conferences

through you know the 48 hour conferences on mind numbing I ideas uh but now if

on mind numbing I ideas uh but now if you look at a lot of the rule making

you look at a lot of the rule making Beijing had a big hand in it because we

Beijing had a big hand in it because we weren't paying attention so I do think

weren't paying attention so I do think it's really important to recognize that

it's really important to recognize that they a lot gets buried in by mattress

they a lot gets buried in by mattress mice and you have to make sure that

mice and you have to make sure that they're your mattress mice mice not

they're your mattress mice mice not somebody else's that are involved in

somebody else's that are involved in those discussions so I'm reformed on

those discussions so I'm reformed on that and and have and have by the way

that and and have and have by the way mentioned it to my successors that maybe

mentioned it to my successors that maybe they should think differently about it

they should think differently about it um so the other another point you talked

um so the other another point you talked about was universities and we've heard a

about was universities and we've heard a lot today about the scale of these AI

lot today about the scale of these AI Foundation models and also as as you

Foundation models and also as as you said you know you have a a new uh thing

said you know you have a a new uh thing with Jennifer Widow at the Hoover

with Jennifer Widow at the Hoover institution on putting Academia in the

institution on putting Academia in the mix but how does Academia get in the mix

mix but how does Academia get in the mix without the without the massive

without the without the massive resources well when you talk about um AI

resources well when you talk about um AI um and I the panel before was just

um and I the panel before was just discussing this you know and and it the

discussing this you know and and it the the problem of not enough uh not enough

the problem of not enough uh not enough compute power um no University or

compute power um no University or combinations of universities can do the

combinations of universities can do the kind of high-scale generative AI large

kind of high-scale generative AI large language models that we're seeing in the

language models that we're seeing in the commercials sector and uh you know

commercials sector and uh you know that's a question I think for the

that's a question I think for the country uh is it the case that we want

country uh is it the case that we want this level of technology to only be done

this level of technology to only be done in uh you know there'll be competition

in uh you know there'll be competition you know Microsoft and and Google and

you know Microsoft and and Google and maybe meta will uh will go to the mat

maybe meta will uh will go to the mat with one another but the um look I'm is

with one another but the um look I'm is has died in the wol capitalist as you'll

has died in the wol capitalist as you'll ever find so let me just put that on the

ever find so let me just put that on the the table but do I really want this only

the table but do I really want this only to be in the commercial sector and I

to be in the commercial sector and I think that's a question for the country

think that's a question for the country to ask and if in fact you would like to

to ask and if in fact you would like to have the capabilities to do some of

have the capabilities to do some of these things one thing maybe there are

these things one thing maybe there are industry collaborations with

industry collaborations with universities that might be one way to

universities that might be one way to think about it or maybe uh this is a

think about it or maybe uh this is a public good in some sense that you want

public good in some sense that you want uh to exist somewhere in the uh

uh to exist somewhere in the uh ecosystem around you know some people

ecosystem around you know some people have talked about National Labs or

have talked about National Labs or whatever uh I'm not smart enough to to

whatever uh I'm not smart enough to to make to know the answer but I am smart

make to know the answer but I am smart enough to ask the question and I think

enough to ask the question and I think we ought to uh we ought to ask that

we ought to uh we ought to ask that question I think it's a question by the

question I think it's a question by the way that uh you hear being uh asked by

way that uh you hear being uh asked by some of the most forward leaning uh of

some of the most forward leaning uh of private sector leaders uh as to what's

private sector leaders uh as to what's the proper um infrastructure

the proper um infrastructure distribution for some of the highest end

distribution for some of the highest end research that might might be done uh we

research that might might be done uh we do know we've had discussions about the

do know we've had discussions about the brain drain um from universities um and

brain drain um from universities um and people think it's you know a lot about

people think it's you know a lot about salary maybe some of it's about salary

salary maybe some of it's about salary but I'm told that a fair amount of it is

but I'm told that a fair amount of it is about can I do the kind of research that

about can I do the kind of research that I want to do uh in a university or do I

I want to do uh in a university or do I have to go and do that in industry and

have to go and do that in industry and that's a that's a big question for the

that's a that's a big question for the country okay um going to shift gears

country okay um going to shift gears here uh a little bit more about National

here uh a little bit more about National Security and and you know I know we

Security and and you know I know we spend a lot of money on cyber security

spend a lot of money on cyber security I'm on the board of the Stanford Credit

I'm on the board of the Stanford Credit Union and we worry about it but you know

Union and we worry about it but you know we're a small piece compared to Chase

we're a small piece compared to Chase Bank or something like that where they

Bank or something like that where they spend a lot more and throughout

spend a lot more and throughout throughout the economy we spend a lot on

throughout the economy we spend a lot on cyber security and I'm sure we're

cyber security and I'm sure we're goingon to have to spend more um that's

goingon to have to spend more um that's all I know whenever the it guys come to

all I know whenever the it guys come to me we need to spend more um and we

me we need to spend more um and we always say yes because they scare us yes

always say yes because they scare us yes um uh so do you

um uh so do you uh kind of bring bringing this question

uh kind of bring bringing this question into a couple things do the advances in

into a couple things do the advances in AI increase your fear and do you think

AI increase your fear and do you think that it's a fear of nation state actors

that it's a fear of nation state actors or what I would call more Rogue actors

or what I would call more Rogue actors that can you know leverage this

that can you know leverage this technology and where where should we be

technology and where where should we be worried and what should we think about

worried and what should we think about well the the big difference um with AI

well the the big difference um with AI um and really all of these Frontier

um and really all of these Frontier Technologies except space where it's a

Technologies except space where it's a mix between uh public and private um is

mix between uh public and private um is that this is kind of the the first time

that this is kind of the the first time that the government hasn't owned the

that the government hasn't owned the domain in which the competition is

domain in which the competition is taking place so uh if you go back to

taking place so uh if you go back to nuclear for instance um I you know there

nuclear for instance um I you know there are always the stories you know the kid

are always the stories you know the kid who reads on the internet how to make a

who reads on the internet how to make a nuclear weapon and holds some City

nuclear weapon and holds some City hostage it's hard to make a nuclear

hostage it's hard to make a nuclear weapon right it's really hard that's the

weapon right it's really hard that's the one of the

one of the reasons countries keep failing at it so

reasons countries keep failing at it so I never actually worried about that

I never actually worried about that but some of the things that are floating

but some of the things that are floating out there uh in the um private domain

out there uh in the um private domain that is some combination of cyber and Ai

that is some combination of cyber and Ai and now it looks as if those Rogue

and now it looks as if those Rogue actors might be more of a problem

actors might be more of a problem they've certainly been a problem in

they've certainly been a problem in cyber but I still think the biggest

cyber but I still think the biggest problem is the state actor um I've

problem is the state actor um I've always been much more concerned about

always been much more concerned about the um how much um the the state actor

the um how much um the the state actor can bring in terms of resources in terms

can bring in terms of resources in terms of intelligence as to where to target um

of intelligence as to where to target um and the state sponsored Rogue actors so

and the state sponsored Rogue actors so when Colonial pipeline went down it was

when Colonial pipeline went down it was a so-called russal Russian criminal

a so-called russal Russian criminal group that uh apparently we believe did

group that uh apparently we believe did it but something very interesting

it but something very interesting happened that criminal group then

happened that criminal group then disappeared and I don't know whether we

disappeared and I don't know whether we told Vladimir Vladimir Putin disappear

told Vladimir Vladimir Putin disappear them or we will or whether we

them or we will or whether we disappeared them I really don't care but

disappeared them I really don't care but the fact is State actors got involved in

the fact is State actors got involved in that kind of case as well so I worry a

that kind of case as well so I worry a lot more about the power that state

lot more about the power that state actors can can bring to this yes deep

actors can can bring to this yes deep fakes and all of that it's this realm

fakes and all of that it's this realm opens up a lot of possibility for a lot

opens up a lot of possibility for a lot of Bad actors but um State actors I'd go

of Bad actors but um State actors I'd go first after state State actors and and

first after state State actors and and one advantage with State actors is that

one advantage with State actors is that they actually have things to lose and um

they actually have things to lose and um one reason that we have avoided a

one reason that we have avoided a nuclear confrontation is uh you do it to

nuclear confrontation is uh you do it to me I'll do it to you and that I think

me I'll do it to you and that I think has possibly been persuad persuasive in

has possibly been persuad persuasive in some of this higher-end cyber activity

some of this higher-end cyber activity that we kept expecting and haven't

that we kept expecting and haven't gotten so this is uh you know earlier we

gotten so this is uh you know earlier we were talking about sort of The Cutting

were talking about sort of The Cutting Edge a technology in competition with

Edge a technology in competition with China as sort of a developer of

China as sort of a developer of technology and it seems like this is

technology and it seems like this is also now hey this doesn't have to be

also now hey this doesn't have to be somebody who's at The Cutting Edge of

somebody who's at The Cutting Edge of Technology but a user of technology and

Technology but a user of technology and the more we advance it the better things

the more we advance it the better things go the better it is for users of All

go the better it is for users of All Sorts but also these other these other

Sorts but also these other these other actors there will always be some malign

actors there will always be some malign actors there always will be and you know

actors there always will be and you know I I assume that at some point something

I I assume that at some point something bad is going to happen because of a

bad is going to happen because of a maligned actor um and we will do

maligned actor um and we will do everything that we can in every sector

everything that we can in every sector to avoid that but uh human history would

to avoid that but uh human history would suggest that somebody's going to manage

suggest that somebody's going to manage it at some point so in in a minute after

it at some point so in in a minute after the next question I'm going to open it

the next question I'm going to open it to the audience for questions um but I

to the audience for questions um but I we've talked a lot about regulation do

we've talked a lot about regulation do you have you know sort of you you were

you have you know sort of you you were concerned about not over regulating but

concerned about not over regulating but are there certain things that you think

are there certain things that you think we should be doing either in terms of

we should be doing either in terms of Regulation or promoting you know

Regulation or promoting you know research or trying to figure out what do

research or trying to figure out what do we do how do we how do we think about

we do how do we how do we think about that well I would start with let's not

that well I would start with let's not start regulating the research itself

start regulating the research itself right um and you know I I do think that

right um and you know I I do think that there are some Norms that you might

there are some Norms that you might start to try to establish uh that you

start to try to establish uh that you don't want to think do things that would

don't want to think do things that would cause Mass casualties might be uh one

cause Mass casualties might be uh one Norm that you could have uh but I would

Norm that you could have uh but I would take it from that perspective rather

take it from that perspective rather than what do we have to stop because I

than what do we have to stop because I just want to repeat government's not

just want to repeat government's not going to be very good at knowing what to

going to be very good at knowing what to stop and they're going to stop something

stop and they're going to stop something that either was never going to

that either was never going to materialize or something's going to

materialize or something's going to materialized in a completely different

materialized in a completely different re different uh uh way and so I just um

re different uh uh way and so I just um I guess I'm just a let's let's talk

I guess I'm just a let's let's talk about it for a while Let's uh uh have

about it for a while Let's uh uh have these conversations between those who

these conversations between those who are actually doing the Innovation and

are actually doing the Innovation and those who have to worry about the

those who have to worry about the institutions let's have that go on for a

institutions let's have that go on for a while uh we'll learn a lot in that uh

while uh we'll learn a lot in that uh period of time but trying to

period of time but trying to prejudge what might be a problem seems

prejudge what might be a problem seems to me not a very smart way to go about

to me not a very smart way to go about it given how fast it's it's uh emerging

it given how fast it's it's uh emerging giving the wild directions in which it's

giving the wild directions in which it's going um and there's one other point

going um and there's one other point that I would make we will have these

that I would make we will have these discussions in our

discussions in our democracy uh we will have Congressional

democracy uh we will have Congressional hearings if something goes wrong we will

hearings if something goes wrong we will have investigative reporters that are

have investigative reporters that are looking at this that's why I make the

looking at this that's why I make the the the case again that I worry a lot

the the case again that I worry a lot less about this in the hands of

less about this in the hands of democracies than in the hands of

democracies than in the hands of autocracies uh ask yourself the question

autocracies uh ask yourself the question if the uh the Soviet Union or Nazi

if the uh the Soviet Union or Nazi Germany had won the race to nuclear

Germany had won the race to nuclear nuclear weapons

nuclear weapons uh you know we did in fact use them

uh you know we did in fact use them twice uh and scared ourselves enough

twice uh and scared ourselves enough that we thought we didn't want to do

that we thought we didn't want to do that again um so I think that you you

that again um so I think that you you want to make sure that this happens in

want to make sure that this happens in an environment in which the debate is

an environment in which the debate is open great thank you so now I want to

open great thank you so now I want to take questions I see one over here yeah

take questions I see one over here yeah I was Wonder wait wait for the mic

I was Wonder wait wait for the mic because the people online can't hear

because the people online can't hear yeah secretary rice I was wondering if

yeah secretary rice I was wondering if you had any thoughts about what might

you had any thoughts about what might happen between China and Taiwan in the

happen between China and Taiwan in the next couple years yeah sure um well wish

next couple years yeah sure um well wish I do uh I I um I find it hard to believe

I do uh I I um I find it hard to believe having talked to a lot of military

having talked to a lot of military people like this that China plans uh an

people like this that China plans uh an amphibious landing on

amphibious landing on Taiwan I've been told by by milal

Taiwan I've been told by by milal military people we're talking about

military people we're talking about something that makes D-Day look like

something that makes D-Day look like Child's Play right there are lots of

Child's Play right there are lots of other ways to put pressure pressure on

other ways to put pressure pressure on Taiwan and I think we've seen a little

Taiwan and I think we've seen a little hint of it over this last week and we

hint of it over this last week and we saw a little hint of it after Nancy

saw a little hint of it after Nancy Pelosi's visit which is a denial

Pelosi's visit which is a denial strategy something that looks like a

strategy something that looks like a quarantine something that keeps Taiwan

quarantine something that keeps Taiwan which has to be a trading Nation from

which has to be a trading Nation from Trading may be combined with uh cyber

Trading may be combined with uh cyber may be combined with cutting underwater

may be combined with cutting underwater sea cables um and something that is uh

sea cables um and something that is uh salami tactic enough alike that it's

salami tactic enough alike that it's hard for the United States to know when

hard for the United States to know when it needs to intervene and one I have a a

it needs to intervene and one I have a a friend who works does a lot of work at

friend who works does a lot of work at the defense department and this person

the defense department and this person says that one thing that worries uh her

says that one thing that worries uh her is that every time she goes back to a

is that every time she goes back to a defense department they're talking about

defense department they're talking about the amphibious landing and these other

the amphibious landing and these other uh ways of of really isolating Taiwan

uh ways of of really isolating Taiwan and making life difficult for Taiwan

and making life difficult for Taiwan they clearly don't like president lii or

they clearly don't like president lii or we see that um but there are lots of

we see that um but there are lots of ways for China to put pressure on Taiwan

ways for China to put pressure on Taiwan that might make a lot more sense and I

that might make a lot more sense and I hope we're beginning to think through

hope we're beginning to think through how we deter that kind of behavior

how we deter that kind of behavior rather than thinking how we deter an

rather than thinking how we deter an allout

invasion may do one here and then go over to the other

over to the other side secretary R I'm Josh glck I'm a

side secretary R I'm Josh glck I'm a visiting scholar at seesac here at

visiting scholar at seesac here at Stanford you talked about the dynamic

Stanford you talked about the dynamic that exists today whereas the bulk of

that exists today whereas the bulk of innovation is happening in the

innovation is happening in the commercial sector whereas you know Cold

commercial sector whereas you know Cold War era and before the government was

War era and before the government was driving much of the technological

driving much of the technological innovation so I'm wondering how you

innovation so I'm wondering how you think the US government is doing today

think the US government is doing today with respect to being able to leverage

with respect to being able to leverage the commercial sector to acquire the

the commercial sector to acquire the technologies that are clearly going to

technologies that are clearly going to be needed for national security purposes

be needed for national security purposes especially as we look at China with a

especially as we look at China with a very different model where they're

very different model where they're essentially able to kind of direct their

essentially able to kind of direct their priorities into their private companies

priorities into their private companies for National Security development uh

for National Security development uh thank you and I think we're doing very

thank you and I think we're doing very poorly very poorly and it's largely

poorly very poorly and it's largely because the procurement and acquisition

because the procurement and acquisition processes in the defense department are

processes in the defense department are broke to say the least and no small

broke to say the least and no small company that has a brilliant young a

company that has a brilliant young a brilliant idea or a brilliant product

brilliant idea or a brilliant product that could make uh that would be lower

that could make uh that would be lower cost and more efficient uh almost none

cost and more efficient uh almost none of them can penetrate the

of them can penetrate the Pentagon and there have been multiple

Pentagon and there have been multiple people that are trying to deal with with

people that are trying to deal with with this problem I know seesac talks about

this problem I know seesac talks about this all the time we we have these uh

this all the time we we have these uh what we call uh track twos between the

what we call uh track twos between the private sector and and the government

private sector and and the government because they don't speak the same

because they don't speak the same language and uh you know we had uh diu

language and uh you know we had uh diu out here that Ash Carter started and uh

out here that Ash Carter started and uh you know the Pentagon has kind of tried

you know the Pentagon has kind of tried parts of it to kill it in its crib from

parts of it to kill it in its crib from time to time and we just keep doing this

time to time and we just keep doing this in a way and then there's the you know

in a way and then there's the you know we should just make it inside or we

we should just make it inside or we should give it to one of the big guys

should give it to one of the big guys and let them make it and we are really

and let them make it and we are really uh not taking advantage of what is our

uh not taking advantage of what is our enormous Advantage which is the the

enormous Advantage which is the the distributed Innovation that our system

distributed Innovation that our system produces and we ought to be snapping up

produces and we ought to be snapping up these Innovations and taking them to the

these Innovations and taking them to the field as fast as possible and part of

field as fast as possible and part of this is a congressional problem now

this is a congressional problem now Congress uh makes us extremely sometimes

Congress uh makes us extremely sometimes risk averse about these things

risk averse about these things uh but sometimes you have to try things

uh but sometimes you have to try things and it may not completely work out but

and it may not completely work out but you know when we put one of those

you know when we put one of those enormous satellites into uh into space

enormous satellites into uh into space uh which is billions of dollars over

uh which is billions of dollars over budget and years behind schedule nobody

budget and years behind schedule nobody said oh boy that was a failure so we

said oh boy that was a failure so we shouldn't do that again so why not take

shouldn't do that again so why not take some of these smaller risks so I think

some of these smaller risks so I think it's a big problem for us it's something

it's a big problem for us it's something we really need to work just a very quick

we really need to work just a very quick story about it so when I was trying to

story about it so when I was trying to do some something that's very much in

do some something that's very much in the pce he was trying to do Israeli

the pce he was trying to do Israeli Palestinian issues and uh we had a

Palestinian issues and uh we had a settlement freeze in place with Israel

settlement freeze in place with Israel and they would come and they would say

and they would come and they would say oh no we're not we're not building

oh no we're not we're not building settlements and I knew from uh from our

settlements and I knew from uh from our nro uh that actually we could see

nro uh that actually we could see certain things happening but if I'd been

certain things happening but if I'd been waiting for clearance to show any of

waiting for clearance to show any of that to the Israelis it would have taken

that to the Israelis it would have taken forever so this is there was this thing

forever so this is there was this thing called Google Earth

called Google Earth and so I took out Google Earth and I

and so I took out Google Earth and I said look at this look how that is going

said look at this look how that is going that way if it's going that way it's

that way if it's going that way it's okay but and so every month we had a

okay but and so every month we had a Google Earth test all right so there's

Google Earth test all right so there's so much in the private sector that could

so much in the private sector that could be helping us and I don't think we're

be helping us and I don't think we're making taking advantage of

making taking advantage of it

okay uh secretary Bryce I'm David Lowry from the University of Georgia music

from the University of Georgia music business program it's a little off topic

business program it's a little off topic but kind of not um I share your views on

but kind of not um I share your views on free markets and deregulation and such

free markets and deregulation and such I'm the only one in the music business

I'm the only one in the music business perhaps but um um do there's sort of

perhaps but um um do there's sort of this notion that we need to basically

this notion that we need to basically sort of suspend the property rights of

sort of suspend the property rights of songwriters and musicians so that we can

songwriters and musicians so that we can ingest a lot of this and train AI

ingest a lot of this and train AI models um that and then essentially that

models um that and then essentially that would compete with us for

would compete with us for Rue however one of the great Innovations

Rue however one of the great Innovations of the American music industry has been

of the American music industry has been our contributions to soft power yeah do

our contributions to soft power yeah do you have any thoughts on that yeah you

you have any thoughts on that yeah you know it's interesting I I don't know if

know it's interesting I I don't know if you know but I was actually a music

you know but I was actually a music major in college and so this is

major in college and so this is something I've actually followed because

something I've actually followed because I'm just sort of interested um and you

I'm just sort of interested um and you know the whole question about IP and

know the whole question about IP and music goes back a long long time right

music goes back a long long time right one of the things that we know was that

one of the things that we know was that Mozart died penus but uh Mozart's wife

Mozart died penus but uh Mozart's wife did not because she actually fought for

did not because she actually fought for the the IP for U the rights to Mozart's

the the IP for U the rights to Mozart's music so it's been going on for a while

music so it's been going on for a while that's my point um look I think this is

that's my point um look I think this is a whole area one of the things that AI

a whole area one of the things that AI is doing is it's challenging just about

is doing is it's challenging just about every area of our uh societal and legal

every area of our uh societal and legal framework um and I I think that we need

framework um and I I think that we need to start to look at in those terms so

to start to look at in those terms so it's not we don't want to rush to a

it's not we don't want to rush to a conclusion about what AI uh sweeping the

conclusion about what AI uh sweeping the net and therefore creating something

net and therefore creating something that was really taken from the IP of

that was really taken from the IP of others and then how do we get proper

others and then how do we get proper compensation to those whose IP it really

compensation to those whose IP it really was you know I don't think we can solve

was you know I don't think we can solve these problems kind of one at a time uh

these problems kind of one at a time uh it really needs a hard look at what the

it really needs a hard look at what the legal framework uh is going to be for

legal framework uh is going to be for these questions around IP because it's

these questions around IP because it's not just in music it's in uh a whole

not just in music it's in uh a whole realm of uh of the creative world where

realm of uh of the creative world where IP is created and then can be uh

IP is created and then can be uh harnessed by uh others without proper

harnessed by uh others without proper compensation and that's really the

compensation and that's really the question uh one of the people that has

question uh one of the people that has just come to to Hoover is a a lawyer

just come to to Hoover is a a lawyer named Eugene volik who is um from uh

named Eugene volik who is um from uh he's a was a professor at UCLA he's a

he's a was a professor at UCLA he's a First Amendment uh lawyer but very

First Amendment uh lawyer but very interested in a lot of these questions

interested in a lot of these questions about how technology is changing and

about how technology is changing and transforming uh the needs of our legal

transforming uh the needs of our legal structure and we've we've had these

structure and we've we've had these challenges to it before but I think a

challenges to it before but I think a kind of Reason we've got to get the

kind of Reason we've got to get the lawyers not to litigate as much as maybe

lawyers not to litigate as much as maybe get some of the lawyers who might

get some of the lawyers who might actually be able to think through these

actually be able to think through these things constitutionally and

things constitutionally and philosophically first um I think that

philosophically first um I think that would be a very helpful and important

would be a very helpful and important set of conversations to have I I now you

set of conversations to have I I now you guys have had a chance to see why I was

guys have had a chance to see why I was so impressed with Dr Rice in front of a

so impressed with Dr Rice in front of a classroom now she's been in front of you

classroom now she's been in front of you she's done this a couple of times before

she's done this a couple of times before so she may be a little practice at it

so she may be a little practice at it but but but fantastic and sort of wide

but but but fantastic and sort of wide ranging from AI to competition to music

ranging from AI to competition to music and everything and I want to thank you

and everything and I want to thank you very much for being here today thank

great you let's take a 10 minutes break and come back at

am I good all right everyone uh if you could take your seats we're going to get

could take your seats we're going to get started with the next uh fireside chat

started with the next uh fireside chat on the

on the agenda we will continue our conversation

agenda we will continue our conversation about intellectual property and the

about intellectual property and the intersection with artificial

intersection with artificial intelligence um we have the next

intelligence um we have the next fireside chat focusing on uh balancing

fireside chat focusing on uh balancing Creator rights and competition and we I

Creator rights and competition and we I would like to welcome to the stage um uh

would like to welcome to the stage um uh two two uh uh great uh friends of the

two two uh uh great uh friends of the antitrust division one that is the

antitrust division one that is the principal uh Deputy Assistant Attorney

principal uh Deputy Assistant Attorney General um for the an trust division

General um for the an trust division Doan mcky and director Kathy vdal from

Doan mcky and director Kathy vdal from the US uh PTO I'm pleased to introduce

the US uh PTO I'm pleased to introduce both of them um I'll let them take a

both of them um I'll let them take a seat here director Kathy Vidal has been

seat here director Kathy Vidal has been the the under Secretary of Commerce for

the the under Secretary of Commerce for intellectual property and the director

intellectual property and the director of the uh patent and trademark office

of the uh patent and trademark office since uh 2021 during her tenure director

since uh 2021 during her tenure director bodal has been focused on incentivizing

bodal has been focused on incentivizing and protecting us Innovation uh

and protecting us Innovation uh entrepreneurship and creativity and she

entrepreneurship and creativity and she has been a partner in uh our mission in

has been a partner in uh our mission in carrying out the administration's whole

carrying out the administration's whole government approach to promoting

government approach to promoting competition in the US economy I I would

competition in the US economy I I would also like to uh tell you a little bit

also like to uh tell you a little bit more about principal Deputy Assistant

more about principal Deputy Assistant Attorney General uh for the antitrust

Attorney General uh for the antitrust division Doha mecki uh Doha has served

division Doha mecki uh Doha has served in this role since uh 2022 she's

in this role since uh 2022 she's responsible for overseeing the

responsible for overseeing the division's Civil enforcement criminal

division's Civil enforcement criminal enforcement litigation uh domestic and

enforcement litigation uh domestic and international policy advocacy our expert

international policy advocacy our expert analysis group and just about every

analysis group and just about every aspect of the division's work uh so

aspect of the division's work uh so we're very pleased to to have here have

we're very pleased to to have here have her here um and Doha before that was a

her here um and Doha before that was a uh a trial attorney who led a number of

uh a trial attorney who led a number of Investigations for the antitrust

Investigations for the antitrust division including those involving labor

division including those involving labor market so I will leave it to DOA thank

market so I will leave it to DOA thank you okay um it is really such a treat to

you okay um it is really such a treat to be here and to get to interview um my

be here and to get to interview um my friend uh Kathy Vidal who um the one

friend uh Kathy Vidal who um the one thing that was left out of uh her bio is

thing that was left out of uh her bio is that she is the real deal um she is uh

that she is the real deal um she is uh the

the administration's uh you know top adviser

administration's uh you know top adviser on all things IP and so it's really

on all things IP and so it's really great um that we get to have her here um

great um that we get to have her here um Kathy many people um may not be aware of

Kathy many people um may not be aware of the really strong

the really strong interconnectedness of Ip and competition

interconnectedness of Ip and competition and so I wonder uh can you tell us a

and so I wonder uh can you tell us a little a little bit about how you see

little a little bit about how you see the interplay between the US IP system

the interplay between the US IP system and fostering a competitive

and fostering a competitive Marketplace so first of all it's really

Marketplace so first of all it's really great to be here this has been so

great to be here this has been so amazing the entire conference so thank

amazing the entire conference so thank you doj thank you Stanford for all of

you doj thank you Stanford for all of this uh I will say that one thing that

this uh I will say that one thing that you've heard over and over again in this

you've heard over and over again in this conference on competition is it's about

conference on competition is it's about Innovation and it's about the economy

Innovation and it's about the economy and so that is exactly what IP is about

and so that is exactly what IP is about if we don't have a strong IP system we

if we don't have a strong IP system we don't have New Market entrance because

don't have New Market entrance because they don't have the market power they

they don't have the market power they need the intellectual property and in

need the intellectual property and in order to get funding and in order to

order to get funding and in order to compete and so when we think about in

compete and so when we think about in the Department of Commerce because I'm

the Department of Commerce because I'm also a high level official within the

also a high level official within the Department of Commerce when we think

Department of Commerce when we think about the economy and think about

about the economy and think about bringing more people into the economy

bringing more people into the economy and lifting more people we need both IP

and lifting more people we need both IP and competition for that and can you

and competition for that and can you talk a little bit about how you see it

talk a little bit about how you see it in Ai and AI related industry in

in Ai and AI related industry in particular so we've all heard that the

particular so we've all heard that the big concern is that little Tech loses

big concern is that little Tech loses out uh to Big Tech and that's something

out uh to Big Tech and that's something that we're trying to solve for across

that we're trying to solve for across the administration you know certainly we

the administration you know certainly we had the president uh receive the uh

had the president uh receive the uh assurances and and and uh from Private

assurances and and and uh from Private Industry uh we coming up on the two-year

Industry uh we coming up on the two-year anniversary of that in uh in two months

anniversary of that in uh in two months or at least the anniversary of that in

or at least the anniversary of that in two months uh and so really what we need

two months uh and so really what we need to do is make sure that in the AI space

to do is make sure that in the AI space that everybody has access and can use AI

that everybody has access and can use AI to innovate because if they don't have

to innovate because if they don't have access all the other work that we've

access all the other work that we've been doing and we've been doing a lot of

been doing and we've been doing a lot of it to bring more people into the

it to bring more people into the ecosystem to lift innovators to lift

ecosystem to lift innovators to lift entrepreneurs none of that is going to

entrepreneurs none of that is going to take effect so I'd like to talk a little

take effect so I'd like to talk a little bit more about um some things that the

bit more about um some things that the administration is doing uh in that vein

administration is doing uh in that vein so uh this past November President Biden

so uh this past November President Biden uh signed an executive order and it is

uh signed an executive order and it is focused on promoting safe cure and

focused on promoting safe cure and trustworthy development of AI um and I

trustworthy development of AI um and I know that uh PTO which you lead had a

know that uh PTO which you lead had a significant hand in shaping it and um

significant hand in shaping it and um will have a hand in carrying it out I

will have a hand in carrying it out I wonder if you can tell us more about

wonder if you can tell us more about that so it's it's a terribly important

that so it's it's a terribly important executive order and if you've seen it

executive order and if you've seen it it's pages long it's very long um I will

it's pages long it's very long um I will say that the way we Orient to these

say that the way we Orient to these issues is really around opportunity and

issues is really around opportunity and around Innovation so our sister agency

around Innovation so our sister agency nist is working on all the risks they

nist is working on all the risks they get to handle all that um from our

get to handle all that um from our perspective we're trying to figure out

perspective we're trying to figure out how the country moves forward uh in

how the country moves forward uh in competition with China we've heard uh

competition with China we've heard uh that from secretary rice you know how we

that from secretary rice you know how we encourage more innovation in the US in

encourage more innovation in the US in all areas enabled by AI as well as

all areas enabled by AI as well as innovation in the AI space so if you

innovation in the AI space so if you look at the executive order we've given

look at the executive order we've given ourselves a few things that we need to

ourselves a few things that we need to deliver on uh we've already delivered on

deliver on uh we've already delivered on what was originally called inventorship

what was originally called inventorship gu guidance but really it's in the space

gu guidance but really it's in the space of AI assisted Innovation so this is not

of AI assisted Innovation so this is not a this is not invention in the stack

a this is not invention in the stack this is you create something with the

this is you create something with the use of AI what are we going to give you

use of AI what are we going to give you a patent on when is AI being used Too

a patent on when is AI being used Too Much where it could threaten to lock up

Much where it could threaten to lock up Innovation because you can imagine and

Innovation because you can imagine and my my counterpart in Australia said they

my my counterpart in Australia said they had an experiment in the back room where

had an experiment in the back room where they basically set AI loose on the

they basically set AI loose on the problem of create every chair you can

problem of create every chair you can imagine now you could certainly then

imagine now you could certainly then have ai patent all of that in which case

have ai patent all of that in which case we were locking up Innovation and we're

we were locking up Innovation and we're not providing the competition through uh

not providing the competition through uh through the intellectual property

through the intellectual property process so when we created the guidance

process so when we created the guidance on that to determine what we are going

on that to determine what we are going to patent and what we are not going to

to patent and what we are not going to patent we focused on the human

patent we focused on the human contribution that if it's too much AI

contribution that if it's too much AI then it it could lock up Innovation

then it it could lock up Innovation really we want to reward Innovation we

really we want to reward Innovation we want to encourage people to use

want to encourage people to use innovation for opportunity uh but we

innovation for opportunity uh but we want to make sure that a human has

want to make sure that a human has sufficient involvement in that process

sufficient involvement in that process so this is really fascinating to me I

so this is really fascinating to me I had not appreciated before um stepping

had not appreciated before um stepping into this role how much guidance your

into this role how much guidance your office um really shares with the public

office um really shares with the public and the inventorship guidance in

and the inventorship guidance in particular um is fascinating because you

particular um is fascinating because you do see the words quality and human

do see the words quality and human centered and I wonder if you can say a

centered and I wonder if you can say a little bit more about that because those

little bit more about that because those seem like intentional value choices so

seem like intentional value choices so they are so we do as an Administration

they are so we do as an Administration we're focused on human centered right

we're focused on human centered right when we look at AI everything that we do

when we look at AI everything that we do and everything we we think about is

and everything we we think about is human- centered in terms of quality you

human- centered in terms of quality you that's also something that we talk about

that's also something that we talk about a lot we also talk about robust and

a lot we also talk about robust and reliable we want if you secure IP in the

reliable we want if you secure IP in the country we want you be able to use that

country we want you be able to use that IP to get funding we want to make sure

IP to get funding we want to make sure that that IP withstands challenges so

that that IP withstands challenges so that's our end goal with all of it um

that's our end goal with all of it um and then it's really also about not just

and then it's really also about not just you know having the law I'm a lawyer as

you know having the law I'm a lawyer as well but it's not just having the

well but it's not just having the lawyers in the room and trying to figure

lawyers in the room and trying to figure out what the legal response is but where

out what the legal response is but where do we want the country to be like what

do we want the country to be like what do we need to incentivize what do we

do we need to incentivize what do we want people to be doing how do we

want people to be doing how do we incentivize the Investments once the

incentivize the Investments once the Innovation is created so if you look at

Innovation is created so if you look at the ultimate guidance we were able to

the ultimate guidance we were able to luckily ground it in the law which is

luckily ground it in the law which is always helpful we didn't have to step up

always helpful we didn't have to step up step out on a limb uh but it really

step out on a limb uh but it really landed where we think it should land to

landed where we think it should land to advance the country in terms of

advance the country in terms of innovation policy your office is also

innovation policy your office is also working with the copyright office yes to

working with the copyright office yes to do a study of copyright law and AI again

do a study of copyright law and AI again this is hugely interesting for all of us

this is hugely interesting for all of us nerdy lawyers in the room like I'm I'm

nerdy lawyers in the room like I'm I'm just a country antitrust lawyer um and

just a country antitrust lawyer um and so looking at adjacent fields to um see

so looking at adjacent fields to um see what what you guys are um doing to think

what what you guys are um doing to think about these Frontier Technologies is

about these Frontier Technologies is really interesting and so I wonder if

really interesting and so I wonder if you can tell us more about that work and

you can tell us more about that work and how that study is going uh so I will say

how that study is going uh so I will say just to step back one of my many roles

just to step back one of my many roles is I am the adviser to the president on

is I am the adviser to the president on all things intellectual property uh it

all things intellectual property uh it was funny I was answering a question the

was funny I was answering a question the other day and it says who's your second

other day and it says who's your second line supervisor I was on the phone with

line supervisor I was on the phone with somebody and I was like the president

somebody and I was like the president she's like the president of

she's like the president of what the United States um so part part

what the United States um so part part of what the EO requires is I've got to

of what the EO requires is I've got to set forth the USPTO has to set forth

set forth the USPTO has to set forth policy for the administration as it

policy for the administration as it relates to copyright now we have a

relates to copyright now we have a separate copyright office and we work

separate copyright office and we work hand inand on this uh they actually

hand inand on this uh they actually issued a request for comment or they

issued a request for comment or they call it request for information to get

call it request for information to get feedback from stakeholders we're

feedback from stakeholders we're digesting the comments they're digesting

digesting the comments they're digesting the comments they're going to be putting

the comments they're going to be putting out reports and suggesting action we are

out reports and suggesting action we are going to be advising the administration

going to be advising the administration on where we should land on these various

on where we should land on these various issues and with copyright it's about the

issues and with copyright it's about the training uh that we heard about earlier

training uh that we heard about earlier and I'm sure you'll hear more about uh

and I'm sure you'll hear more about uh it's also about the output it's about

it's also about the output it's about name image likeness which is something

name image likeness which is something that we've been leaning in on when it

that we've been leaning in on when it comes to sports because we know that

comes to sports because we know that disproportionately impacts certain

disproportionately impacts certain populations so we've generally been

populations so we've generally been leaning in and we're going to lean in

leaning in and we're going to lean in when it comes to copyright as well uh

when it comes to copyright as well uh and it's about transparency so we're

and it's about transparency so we're going to be setting the policy for the

going to be setting the policy for the administration on all of that while

administration on all of that while working with and you heard the term

working with and you heard the term already like-minded countries uh which

already like-minded countries uh which we're doing right now with the UK um

we're doing right now with the UK um with EU uh and with other countries as

with EU uh and with other countries as well in other areas and can you tell us

well in other areas and can you tell us a little bit more about how in the

a little bit more about how in the course of doing that work you seek to

course of doing that work you seek to balance um you know the need for access

balance um you know the need for access to training data with uh IP protected

to training data with uh IP protected work so it's it's really important and

work so it's it's really important and it's interesting because I didn't

it's interesting because I didn't realize that this before I got into

realize that this before I got into government but when we ask people for

government but when we ask people for their comments they're pretty binary uh

their comments they're pretty binary uh we face this issue with doj when it came

we face this issue with doj when it came to standard essential policy when we

to standard essential policy when we came on board everybody needs to

came on board everybody needs to advocate for what's best for their

advocate for what's best for their company right that's great but where we

company right that's great but where we really solve things is when we get

really solve things is when we get people in a room and say it we're not

people in a room and say it we're not going to come up with a binary answer

going to come up with a binary answer it's not going to be all one person or

it's not going to be all one person or all somebody else we need what's best

all somebody else we need what's best for the country and so when it comes to

for the country and so when it comes to solving this we do read the comments

solving this we do read the comments they're really important but we think

they're really important but we think about not picking winners and losers but

about not picking winners and losers but how can everybody win how can we create

how can everybody win how can we create an ecosystem that doesn't disrupt the

an ecosystem that doesn't disrupt the the literary the music industry you know

the literary the music industry you know Productions like everything that really

Productions like everything that really makes our country beautiful how can we

makes our country beautiful how can we not disrupt that but still allow for the

not disrupt that but still allow for the Innovation that you need based on

Innovation that you need based on ingestion of you know of data so um so

ingestion of you know of data so um so that's what the balance we're trying to

that's what the balance we're trying to strike it's not an easy answer um we are

strike it's not an easy answer um we are trying to attack things as fast as we

trying to attack things as fast as we can it's imperative that we do so we are

can it's imperative that we do so we are cranking out even more guidance that

cranking out even more guidance that you'll see coming up soon on the patent

you'll see coming up soon on the patent side side uh on the copyright side it's

side side uh on the copyright side it's more complicated um and we do need to

more complicated um and we do need to also as we're solving for that make sure

also as we're solving for that make sure we're having those discussions

we're having those discussions internationally because what we don't

internationally because what we don't want is to drive the artists out of the

want is to drive the artists out of the US into other countries because they

US into other countries because they don't feel like they're properly treated

don't feel like they're properly treated in the US same thing with Innovation

in the US same thing with Innovation it's fascinating

it's fascinating um can you tell us what other policy

um can you tell us what other policy related initiatives you guys are

related initiatives you guys are pursuing to promote competition so one

pursuing to promote competition so one of the things that we're pursuing right

of the things that we're pursuing right now is what's called patent eligibility

now is what's called patent eligibility guidance so this is I told you that

guidance so this is I told you that we're we've already put out guidance on

we're we've already put out guidance on whether or not we'll issue a patent

whether or not we'll issue a patent based on who contributes and the extent

based on who contributes and the extent to which AI contributes there's also uh

to which AI contributes there's also uh a different part of the law on what kind

a different part of the law on what kind of things are we going to actually issue

of things are we going to actually issue a patent on so we don't issue patents on

a patent on so we don't issue patents on scientific formulas like there's certain

scientific formulas like there's certain things we won't issue patents on so we

things we won't issue patents on so we are in the process of issuing guidance

are in the process of issuing guidance on AI as related to that what kind of

on AI as related to that what kind of patents in the stack are we going to be

patents in the stack are we going to be issuing patents on what kind of AI

issuing patents on what kind of AI assisted Innovations are we going to be

assisted Innovations are we going to be issuing patents on so that's next um I

issuing patents on so that's next um I will say just while I have a chance on

will say just while I have a chance on the inventorship guidance uh we are

the inventorship guidance uh we are going to reopen the window for people to

going to reopen the window for people to comment so if you didn't know that this

comment so if you didn't know that this existed you can email me directly at

existed you can email me directly at director uspto.gov I'll make sure you

director uspto.gov I'll make sure you have a link we're going to reopen the

have a link we're going to reopen the comment period for a very short period

comment period for a very short period uh because some people file their

uh because some people file their comments late which happens often so

comments late which happens often so you'll have a chance to comment and we

you'll have a chance to comment and we are also seeking comments on other

are also seeking comments on other aspects of patentability so that is open

aspects of patentability so that is open right now uh I could go into a bunch of

right now uh I could go into a bunch of nerdy topics on prior art and all these

nerdy topics on prior art and all these other terms of art um I will say

other terms of art um I will say hopefully it explains it enough in the

hopefully it explains it enough in the uh in the document itself but we want to

uh in the document itself but we want to hit every aspect of patenting because

hit every aspect of patenting because that's where that's really the IP that

that's where that's really the IP that relates to Innovation um and really

relates to Innovation um and really where we feel like we can get

where we feel like we can get stakeholder input and solve for that

stakeholder input and solve for that sooner rather than later so when we were

sooner rather than later so when we were um thinking about this conversation and

um thinking about this conversation and one of the topics we stumbled upon was

one of the topics we stumbled upon was your emerging technology partnership um

your emerging technology partnership um this is hugely interesting I wonder if

this is hugely interesting I wonder if you can give us some background on it

you can give us some background on it and how you're thinking about AI related

and how you're thinking about AI related issues there so first of all kudos to my

issues there so first of all kudos to my team who's probably watching from the DC

team who's probably watching from the DC area they're amazing on the AI front so

area they're amazing on the AI front so uh we founded in June of 2022 uh so

uh we founded in June of 2022 uh so right after I came on board an AI

right after I came on board an AI emerging Technologies part ship because

emerging Technologies part ship because there's a lot we need to solve for and

there's a lot we need to solve for and we want to hear from the public we want

we want to hear from the public we want to hear what you want us to solve for

to hear what you want us to solve for and we want to get your feedback through

and we want to get your feedback through the process so instead of just going

the process so instead of just going through the comment process which you

through the comment process which you know everything we're doing is with data

know everything we're doing is with data we're very datadriven uh we want to

we're very datadriven uh we want to shape what we even should be thinking

shape what we even should be thinking about so through the AI partnership

about so through the AI partnership we've been focused on that uh we focus

we've been focused on that uh we focus mostly on AI to date I will say we got

mostly on AI to date I will say we got to get on Quantum uh before we're behind

to get on Quantum uh before we're behind uh when it comes to Quantum uh but when

uh when it comes to Quantum uh but when it comes to that we are thinking about

it comes to that we are thinking about who can we hear from and different parts

who can we hear from and different parts of the ecosystem to make sure that we

of the ecosystem to make sure that we are creating an IP infrastructure and an

are creating an IP infrastructure and an IP system that encourages competition

IP system that encourages competition right because IP can both discourage

right because IP can both discourage competition or can encourage it and we

competition or can encourage it and we need to make sure we have a pro

need to make sure we have a pro competition pro-innovation IP strategy

competition pro-innovation IP strategy so you guys have done listening sessions

so you guys have done listening sessions as part of this um um partnership um I

as part of this um um partnership um I wonder if you can share with us some of

wonder if you can share with us some of the things that you've heard uh in these

the things that you've heard uh in these listening sessions so I will say that a

listening sessions so I will say that a lot of what we hear you know we've heard

lot of what we hear you know we've heard it because we followed it with action so

it because we followed it with action so if you look at the request for comments

if you look at the request for comments that we put out where we saw specific

that we put out where we saw specific feedback on specific proposals it was

feedback on specific proposals it was because we heard in those sessions what

because we heard in those sessions what the public wanted us to solve for um and

the public wanted us to solve for um and so and it wasn't enough information and

so and it wasn't enough information and we wanted to reach out more broadly once

we wanted to reach out more broadly once we narrowed the issues to make sure we

we narrowed the issues to make sure we had more specific data in just solve for

had more specific data in just solve for them so I'll ask one provocative

them so I'll ask one provocative question uh to end on before we pivot to

question uh to end on before we pivot to our next panel um I think we heard

our next panel um I think we heard someone uh earlier today say that

someone uh earlier today say that government is very good at promoting

government is very good at promoting Innovation um but government is very bad

Innovation um but government is very bad at innovating

at innovating wrong should I end on that no so so

wrong should I end on that no so so first of all I actually have an AI

first of all I actually have an AI background so I was in AI in the 90s

background so I was in AI in the 90s which may if you're in the AI space now

which may if you're in the AI space now you may not think that qualifies that's

you may not think that qualifies that's a hipster comment you did it before it

a hipster comment you did it before it was cool I did it before it was cool uh

was cool I did it before it was cool uh story of my life but yeah so I was in AI

story of my life but yeah so I was in AI in the 90s I have an AI background I'm

in the 90s I have an AI background I'm very focused on it um we have been

very focused on it um we have been innovating at the US PTO in the AI space

innovating at the US PTO in the AI space for quite a while uh and in fact when

for quite a while uh and in fact when chat GPT was out there uh the very first

chat GPT was out there uh the very first thing I said is don't say no say not yet

thing I said is don't say no say not yet and then we immediately established an

and then we immediately established an AI lab so we are testing things with

AI lab so we are testing things with chat GPT type functionality GPT

chat GPT type functionality GPT functionality in the sophisticated

functionality in the sophisticated architecture you need in the office

architecture you need in the office right now on various things we want to

right now on various things we want to roll out and we've been using Ai and

roll out and we've been using Ai and being been a very rapid adopter of AI

being been a very rapid adopter of AI when we can solve for critical needs

when we can solve for critical needs within the agency and I'll say one last

within the agency and I'll say one last thing we have a very excited kaggle

thing we have a very excited kaggle competition right now if you really want

competition right now if you really want to nerd out on AI where we are asking

to nerd out on AI where we are asking the public to solve for how can you

the public to solve for how can you explain the result of AI because if you

explain the result of AI because if you can explain result then you have trust

can explain result then you have trust and so we did do a patent use case on

and so we did do a patent use case on that um this is our second kaggle

that um this is our second kaggle competition and I have to read this this

competition and I have to read this this is from a kaggle master and the top .1%

is from a kaggle master and the top .1% of kaggle contestants he said I think

of kaggle contestants he said I think this is one of the most fascinating

this is one of the most fascinating competitions I've ever seen in kaggle

competitions I've ever seen in kaggle I'm ready to join so we are

innovating okay um that concludes uh this

this conversation um thank you so much Kathy

thank you so much for uh letting us know all the exciting work that the PTO is

all the exciting work that the PTO is doing in AI I'd like now to welcome our

doing in AI I'd like now to welcome our panelists up to continue the discussion

panelists up to continue the discussion of balancing creators rights and

of balancing creators rights and competition and artificial intelligence

competition and artificial intelligence so I I'll welcome the panelists and and

so I I'll welcome the panelists and and uh introduce you all and then turn

uh introduce you all and then turn things over to

right stuck behind doing a little

behind doing a little chuff all right so we have uh Duncan uh

chuff all right so we have uh Duncan uh Crabtree uh Ireland um who is the

Crabtree uh Ireland um who is the national executive director and chief

national executive director and chief negotiator for sagaa uh next to Doan uh

negotiator for sagaa uh next to Doan uh to his left is Professor David Lowry

to his left is Professor David Lowry senior lecturer of music business at the

senior lecturer of music business at the University of Georgia and then we have

University of Georgia and then we have Michael fricklas who is the chief legal

Michael fricklas who is the chief legal officer and corporate Secretary of

officer and corporate Secretary of advanced Publications and we do have

advanced Publications and we do have someone uh uh uh joining us by uh video

someone uh uh uh joining us by uh video although he he just stepped away but I

although he he just stepped away but I know he's coming

know he's coming back uh Jonathan uh taplin who is the

back uh Jonathan uh taplin who is the director am Meritus at The annenburg

director am Meritus at The annenburg Innovation Lab at the University of uh

Innovation Lab at the University of uh of Southern California here in

of Southern California here in California and I will turn things over

California and I will turn things over to Doha to start the conversation thank

to Doha to start the conversation thank you and to to uh director Vidal who's

you and to to uh director Vidal who's also going to be moderating our panel so

also going to be moderating our panel so thank you for doing that for us okay so

thank you for doing that for us okay so today we've discussed a lot of really

today we've discussed a lot of really interesting um topics related to Ai and

interesting um topics related to Ai and AI related Industries and um as an

AI related Industries and um as an antitrust enforcer um as government

antitrust enforcer um as government officials it's really our job to better

officials it's really our job to better understand how these markets work um all

understand how these markets work um all the way from the chip to the consumer uh

the way from the chip to the consumer uh as we like to say

as we like to say um and for the antitrust enforcers in

um and for the antitrust enforcers in the room um we are again seeking to

the room um we are again seeking to understand but also recognizing Our

understand but also recognizing Our obligation to appropriately police um

obligation to appropriately police um aspects of AI markets that raise

aspects of AI markets that raise antitrust concerns and so when we think

antitrust concerns and so when we think about um potential risk to competition

about um potential risk to competition from the development of foundation

from the development of foundation models um that can happen at that level

models um that can happen at that level of the stack but we also know that

of the stack but we also know that control of AI input can either cement or

control of AI input can either cement or entrench dominance of certain

entrench dominance of certain incumbents um but what about the impact

incumbents um but what about the impact on the sources of that training data um

on the sources of that training data um itself um and specifically what about

itself um and specifically what about the people whose talent and labor um

the people whose talent and labor um really is the source of that training

really is the source of that training data um and so that's what this panel is

data um and so that's what this panel is about um we have a number of creators

about um we have a number of creators and um innovators who are going to help

and um innovators who are going to help us us explore um the intersection of

us us explore um the intersection of training data that may be protected by

training data that may be protected by uh intellectual property rights and the

uh intellectual property rights and the need to protect incentives to continue

need to protect incentives to continue to create uh that uh that um IP

to create uh that uh that um IP protected um training data um and so to

protected um training data um and so to maximize time we won't go through

maximize time we won't go through further introductions um I encourage you

further introductions um I encourage you to look through their bios because

to look through their bios because they're these people are fascinating um

they're these people are fascinating um there is someone here who might have

there is someone here who might have uh maybe managed Bob Dylan's uh first

uh maybe managed Bob Dylan's uh first tour and helped uh uh one uh

tour and helped uh uh one uh scorey um produce his first uh major

scorey um produce his first uh major film um and lots of other interesting

film um and lots of other interesting factoids so this is a really great set

factoids so this is a really great set um so to all of our panelists and we'll

um so to all of our panelists and we'll just go down the line um I have two

just go down the line um I have two questions uh that I will ask you guys to

questions uh that I will ask you guys to um each answer in about 3 minutes um

um each answer in about 3 minutes um what are the top two ways in which AI is

what are the top two ways in which AI is affecting competition in your part of

affecting competition in your part of the media and entertainment industry and

the media and entertainment industry and what are the competition costs and

what are the competition costs and benefits that you think all of us should

benefits that you think all of us should be thinking about um in terms of access

be thinking about um in terms of access to training data and compensating for

to training data and compensating for use of Ip or the use of name image and

use of Ip or the use of name image and likeness so Duncan we'll start with you

likeness so Duncan we'll start with you great thank you so much I appreciate

great thank you so much I appreciate that and very excited to be on this

that and very excited to be on this esteem panel with great panelists and

esteem panel with great panelists and talking about this topic now normally

talking about this topic now normally you probably don't have Union

you probably don't have Union Representatives in antitrust

Representatives in antitrust conversations because as those of you

conversations because as those of you who are antitrust experts know uh there

who are antitrust experts know uh there is a both a statutory and a

is a both a statutory and a non-statutory antitrust exemption for

non-statutory antitrust exemption for collective bargaining and activities

collective bargaining and activities that unions engage in but like

that unions engage in but like everything else AI disrupts the world

everything else AI disrupts the world right so as a union that represents

right so as a union that represents performers and just for those who don't

performers and just for those who don't know who I am I'm the the head and chief

know who I am I'm the the head and chief negotiator of sag aftera which is the

negotiator of sag aftera which is the union that represents professional

union that represents professional performers meaning actors also meaning

performers meaning actors also meaning recording artists and singers we also

recording artists and singers we also represent broadcast journalists and

represent broadcast journalists and basically anybody who works in front of

basically anybody who works in front of a camera or behind a microphone that's

a camera or behind a microphone that's what we do and so you might imagine AI

what we do and so you might imagine AI has been a tremendous concern to

has been a tremendous concern to creative talent in all of the industries

creative talent in all of the industries that we're involved in uh on on the two

that we're involved in uh on on the two fronts that Doha identified and maybe

fronts that Doha identified and maybe some others first of all as far as

some others first of all as far as inputs go um this been tremendously

inputs go um this been tremendously challenging of course as I think

challenging of course as I think everyone knows even though a lot of AI

everyone knows even though a lot of AI companies don't want to admit it it's

companies don't want to admit it it's very clear that people building

very clear that people building foundational models people developing AI

foundational models people developing AI systems have been uh in most cases

systems have been uh in most cases simply taking works that they have found

simply taking works that they have found to be publicly available and using them

to be publicly available and using them to train their models and of course

to train their models and of course those of us who live by intellectual

those of us who live by intellectual property know the fact that something is

property know the fact that something is publicly available does not mean that

publicly available does not mean that it's free for the taking um those are

it's free for the taking um those are two different concepts and uh from the

two different concepts and uh from the point of view of someone who represents

point of view of someone who represents creative Talent the last thing that we

creative Talent the last thing that we think is that the work that our members

think is that the work that our members have put in over years and decades in

have put in over years and decades in creating extraordinary creative uh Works

creating extraordinary creative uh Works should simply be taken and used to

should simply be taken and used to create a whole new business A Whole New

create a whole new business A Whole New Concept and then monetized without any

Concept and then monetized without any consent or compensation so one of the

consent or compensation so one of the things that we fought for in our

things that we fought for in our collective bargaining process last year

collective bargaining process last year you may have seen some news about it we

you may have seen some news about it we had a big strike against the studios and

had a big strike against the studios and streamers was the basic concept of

streamers was the basic concept of consent in our case we call it informed

consent in our case we call it informed consent because we think it needs to be

consent because we think it needs to be more than just a boilerplate signature

more than just a boilerplate signature but consent and fair compensation and

but consent and fair compensation and that needs to be underpinned by the

that needs to be underpinned by the intellectual property laws and of course

intellectual property laws and of course having that um underpinning I think also

having that um underpinning I think also helps create a more level and fair

helps create a more level and fair playing field for companies that do want

playing field for companies that do want to create these kinds of models because

to create these kinds of models because if companies uh if companies can use

if companies uh if companies can use their economic leverage their size to

their economic leverage their size to Simply scrape and take whatever they can

Simply scrape and take whatever they can grab without having to negotiate any

grab without having to negotiate any kind of fair compensation or get

kind of fair compensation or get licenses for it that really does tend to

licenses for it that really does tend to um to run to the benefit of the biggest

um to run to the benefit of the biggest and most advanced companies that have

and most advanced companies that have already staked out space in this area or

already staked out space in this area or who just have huge economic resources on

who just have huge economic resources on the output side is also something that's

the output side is also something that's really really important to our members

really really important to our members I.E it is not okay to take someone's

I.E it is not okay to take someone's face someone's voice someone's um you

face someone's voice someone's um you know performance their image their

know performance their image their likeness and simply use it to create

likeness and simply use it to create something without their permission this

something without their permission this really goes to a very core fundamental

really goes to a very core fundamental part of being a human and having the

part of being a human and having the right to control the words that come out

right to control the words that come out of your mouth and the message that you

of your mouth and the message that you choose to associate yourself with or

choose to associate yourself with or deliver this happened to me last year

deliver this happened to me last year when we were uh after we finally reached

when we were uh after we finally reached an agreement in November with the

an agreement in November with the studios someone created a deep fake

studios someone created a deep fake video of me uh saying false things about

video of me uh saying false things about our contract uh and recommending against

our contract uh and recommending against it that was then distributed on social

it that was then distributed on social media tens of thousands of people saw it

media tens of thousands of people saw it including our members and it wasn't me I

including our members and it wasn't me I had nothing to do with it they had

had nothing to do with it they had scraped video from prior videos of me

scraped video from prior videos of me and they used it to do that so let me

and they used it to do that so let me just tell you if you're not a performer

just tell you if you're not a performer you're not an actor don't think that

you're not an actor don't think that you're immune from the effects of this

you're immune from the effects of this technology and it's impact that it could

technology and it's impact that it could have on you in your career in your

have on you in your career in your workplace in your world it's something

workplace in your world it's something that affects all of us and the fact is

that affects all of us and the fact is when you mentioned name image likeness

when you mentioned name image likeness rights that's a perfect example of an

rights that's a perfect example of an area where we really need the law to

area where we really need the law to catch up with the reality that people

catch up with the reality that people are dealing with and this has been the

are dealing with and this has been the case for a number of years many of our

case for a number of years many of our members have been abused through deep

members have been abused through deep fake pornography they've been Abed

fake pornography they've been Abed through misuse of their image voice and

through misuse of their image voice and likeness and it can apply to everyone so

likeness and it can apply to everyone so there is legislation that's happening

there is legislation that's happening there we've done a lot of collective

there we've done a lot of collective bargaining on this topic uh if I'm being

bargaining on this topic uh if I'm being very uh candid we've had more success in

very uh candid we've had more success in collectively bargaining on the output

collectively bargaining on the output side than on the input side because

side than on the input side because almost every company is scared to limit

almost every company is scared to limit themselves on the input side the

themselves on the input side the training side while the litigation is

training side while the litigation is still going through the system about

still going through the system about what you know fair use copyright relates

what you know fair use copyright relates to training data but the reality is is

to training data but the reality is is both sides with the input and the output

both sides with the input and the output side need to have uh a balance that

side need to have uh a balance that respects individuals and respects human

respects individuals and respects human creativity and I was just at a meeting

creativity and I was just at a meeting two days ago with a whole bunch of

two days ago with a whole bunch of corporate Executives from the tech

corporate Executives from the tech industry and other Industries and as a

industry and other Industries and as a labor person I have to remind them it's

labor person I have to remind them it's so important that we strike this balance

so important that we strike this balance and it's important for all of us because

and it's important for all of us because when companies go across that line um

when companies go across that line um that's when we start to have the kind of

that's when we start to have the kind of mric reactions that actually hamper

mric reactions that actually hamper Innovation and hamper the ability to get

Innovation and hamper the ability to get where we want to go there's a lot of

where we want to go there's a lot of people who are calling this you know the

people who are calling this you know the fourth Industrial Revolution and that's

fourth Industrial Revolution and that's great if it's the fourth Industrial

great if it's the fourth Industrial Revolution but if the concerns of

Revolution but if the concerns of ordinary people and workers are not

ordinary people and workers are not taken into account it's going to be a

taken into account it's going to be a totally different kind of Revolution not

totally different kind of Revolution not one that everyone's going to be excited

one that everyone's going to be excited about so that's why we need to stay

about so that's why we need to stay focused on making sure we achieve that

focused on making sure we achieve that balance and um I think that the the

balance and um I think that the the department and the Anti-Trust folks have

department and the Anti-Trust folks have a really important role to play in that

a really important role to play in that because there is a real risk of this

because there is a real risk of this area becoming too dominant ated by a

area becoming too dominant ated by a small group of companies that have

small group of companies that have outsized power in this area and so

outsized power in this area and so that's something that I hope and believe

that's something that I hope and believe that um the antitrust authorities will

that um the antitrust authorities will stay focused on and uh and we can all

stay focused on and uh and we can all work together to make sure that this new

work together to make sure that this new Industrial Revolution Works in a way

Industrial Revolution Works in a way that's beneficial for everybody thanks

that's beneficial for everybody thanks just before we um go all the way down

just before we um go all the way down the line I want to bring Jonathan into

the line I want to bring Jonathan into the conversation um he is joining us

the conversation um he is joining us remotely um but has a number of

remotely um but has a number of interesting uh insights to share so John

interesting uh insights to share so John Jan can we go to you sure um well look I

Jan can we go to you sure um well look I confess to having had the privilege of

confess to having had the privilege of working with Bob Dylan and Marty scorsi

working with Bob Dylan and Marty scorsi in my younger life before I uh became an

in my younger life before I uh became an academic and so I'm not really worried

academic and so I'm not really worried that AI is going to replace true

that AI is going to replace true artists uh because AI is incapable of

artists uh because AI is incapable of original thought all it can do is

original thought all it can do is compile previous work that has been

compile previous work that has been train on but what AI can do and will do

train on but what AI can do and will do in the immortal words of Steve Bannon is

in the immortal words of Steve Bannon is flood the Zone with

flood the Zone with [Laughter]

[Laughter] there's

there's 110,000 new songs uploaded to Spotify

110,000 new songs uploaded to Spotify every single

every single day imagine that so more than half of

day imagine that so more than half of those are what I call AI clones it's

those are what I call AI clones it's some guy

some guy in his basement taking the latest hit

in his basement taking the latest hit from Taylor Swift or someone else and

from Taylor Swift or someone else and making something that sounds kind of

making something that sounds kind of like it posts it to Spotify and then

like it posts it to Spotify and then employs his army of bots to click on it

employs his army of bots to click on it so he's getting lots of revenue from

so he's getting lots of revenue from that track and of course that subtracts

that track and of course that subtracts revenue from the real artists and it's

revenue from the real artists and it's same thing is happening in the in the

same thing is happening in the in the book publishing business

book publishing business uh you know when a book gets pre-ordered

uh you know when a book gets pre-ordered put up for pre-order and it starts to

put up for pre-order and it starts to get a lot of attention on Amazon within

get a lot of attention on Amazon within two weeks there will be an AI knockoff

two weeks there will be an AI knockoff of that

of that book published before the book actually

book published before the book actually comes out you know there are summaries

comes out you know there are summaries of books that are all around any good

of books that are all around any good book that comes out there will be all

book that comes out there will be all sorts of summaries that are sold for $2

sorts of summaries that are sold for $2 or something again AI generated again

or something again AI generated again taking all the money from the legitimate

taking all the money from the legitimate artist Jane fredman who's a well-known

artist Jane fredman who's a well-known author found a cach of garbage books

author found a cach of garbage books written under her name she had nothing

written under her name she had nothing to do with them and they were put up and

to do with them and they were put up and of course the famous thing of you know

of course the famous thing of you know now chat gbt wrote uh the concluding

now chat gbt wrote uh the concluding book in George

book in George Martin's song Of Ice And Fire series he

Martin's song Of Ice And Fire series he had nothing to do with it they just

had nothing to do with it they just posted it so the second question that

posted it so the second question that you asked is what are the competition

you asked is what are the competition cost and benefits we should be thinking

cost and benefits we should be thinking about I think this the wrong way to

about I think this the wrong way to think about

think about it big

it big Tech um has really had a luxury

Tech um has really had a luxury unaffordable in any other business of

unaffordable in any other business of having all its production inputs for

having all its production inputs for free they' like that business Mark andri

free they' like that business Mark andri said says that's the only way generative

said says that's the only way generative AI will work is you'll if the inputs are

AI will work is you'll if the inputs are for free that they don't have to pay for

for free that they don't have to pay for any of that stuff um the problem with

any of that stuff um the problem with that is that's destroyed the newspaper

that is that's destroyed the newspaper business it almost destroyed the music

business it almost destroyed the music business and now it's hard at work

business and now it's hard at work disrupting the book publishing and the

disrupting the book publishing and the film entertainment

film entertainment business Microsoft open AI Google meta

business Microsoft open AI Google meta will continually use the fiction of fair

will continually use the fiction of fair use as a cover for theft

use as a cover for theft your legal work use of these

your legal work use of these works you know has allowed companies to

works you know has allowed companies to become very big and it hasn't offered

become very big and it hasn't offered compensation to the artists on Whose

compensation to the artists on Whose work they're

working okay um let's continue down the line yes what was the second question

line yes what was the second question Doha um the competition cost and

Doha um the competition cost and benefits okay all right so what I think

benefits okay all right so what I think about when I think about AI the first

about when I think about AI the first thing that comes to mind the first sort

thing that comes to mind the first sort of problem

of problem is when we're talking about competition

is when we're talking about competition I think a lot of times people mean you

I think a lot of times people mean you know you're thinking about Monopoly or

know you're thinking about Monopoly or an antimonopoly right but really more

an antimonopoly right but really more fundamental to that is unfair

fundamental to that is unfair competition I'm not a lawyer I'm a

competition I'm not a lawyer I'm a musician as a mathematician I'm an

musician as a mathematician I'm an academic um I don't know exactly how

academic um I don't know exactly how they're related but I understand unfair

they're related but I understand unfair competition you know in a in a sort of a

competition you know in a in a sort of a layman's terms and that's what we're

layman's terms and that's what we're seeing for

seeing for instance uh there's a number of my

instance uh there's a number of my friends do musical cues for television

friends do musical cues for television and film usually not they're not like

and film usually not they're not like huge film large television stuff like

huge film large television stuff like that and these are the little Little

that and these are the little Little Snips of Music little emotional sort of

Snips of Music little emotional sort of pieces or create tension or release you

pieces or create tension or release you know tension and stuff like that they're

know tension and stuff like that they're finding

finding that

that essentially they've been

essentially they've been scraped

scraped and and then essentially people have

and and then essentially people have sort of these you know basically buy Ami

sort of these you know basically buy Ami music they may not even know that

music they may not even know that they're B buying AI music essentially

they're B buying AI music essentially their own customers are being sold

their own customers are being sold essentially music that's derived from

essentially music that's derived from their own music you know at a cheaper

their own music you know at a cheaper rate right so their own music is being

rate right so their own music is being unfairly used to compete against them

unfairly used to compete against them you also see this with graphic artists I

you also see this with graphic artists I mean it's just that that business like

mean it's just that that business like you know people who are illustrators

you know people who are illustrators create graphic art and stuff like that

create graphic art and stuff like that that business is being destroyed right

that business is being destroyed right now the problem with unfair competition

now the problem with unfair competition you know I mean you could be an

you know I mean you could be an individual and say well that's really

individual and say well that's really bad that really sucks because that made

bad that really sucks because that made my job go away or I don't I have half

my job go away or I don't I have half the revenue that I used to right but

the revenue that I used to right but unfair competition is deeper than that

unfair competition is deeper than that because what it is it's essentially it's

because what it is it's essentially it's essentially allowing Free Riders people

essentially allowing Free Riders people with no skin in the game to make money

with no skin in the game to make money right off of somebody else's work right

right off of somebody else's work right to eat up like a bunch of the revenue I

to eat up like a bunch of the revenue I personally feel that that's like a

personally feel that that's like a really damaging cooning to society that

really damaging cooning to society that we shouldn't tolerate

we shouldn't tolerate right that's the whole point of

right that's the whole point of civilization rule of law games are fair

civilization rule of law games are fair the game's Fair we all play by the rules

the game's Fair we all play by the rules right why would we let something like

right why would we let something like that happen and praise it as some sort

that happen and praise it as some sort of great Innovation

of great Innovation right so there's that and then so I'm

right so there's that and then so I'm sorry what was the second

sorry what was the second question um I think you covered a lot

question um I think you covered a lot well let me let me ask this way are

well let me let me ask this way are there any benefits that you think are

there any benefits that you think are worth highlighting sure but not in the

worth highlighting sure but not in the spammy aspects of

spammy aspects of AI uh cryptography

AI uh cryptography decryption right seeking out weak

decryption right seeking out weak signals like in science medical right

signals like in science medical right materials

materials development there's so many things like

development there's so many things like that like especially in the health field

that like especially in the health field it doesn't you don't need to vacuum up

it doesn't you don't need to vacuum up all the songs and literature in the

all the songs and literature in the world to develop that stuff that's where

world to develop that stuff that's where the real Innovation is and that's that

the real Innovation is and that's that doesn't require violating everybody's

doesn't require violating everybody's copyrights or rights of publicity or

copyrights or rights of publicity or they're faking their voice and stuff

they're faking their voice and stuff like that so I actually do think this is

like that so I actually do think this is interesting right I also you know even

interesting right I also you know even some of the basic AI is like sort of

some of the basic AI is like sort of superpower my ability to like sort of

superpower my ability to like sort of design a a lecture and stuff like that

design a a lecture and stuff like that there are good things out there right

there are good things out there right I'm just afraid that the the adoption

I'm just afraid that the the adoption that's happened so fast and all the AI

that's happened so fast and all the AI companies are pointing to is like look

companies are pointing to is like look how look what a great breakthrough this

how look what a great breakthrough this is look how fast it's being adopted was

is look how fast it's being adopted was because they vacuumed up it's mostly

because they vacuumed up it's mostly like you

like you know you

know you know sort of this lwh hanging fruit of

know sort of this lwh hanging fruit of what's being created by Ai and anyway

what's being created by Ai and anyway yeah I find that hugely fascina I'm just

yeah I find that hugely fascina I'm just moderating um but um I hear you to be

moderating um but um I hear you to be saying um um there's a difference

saying um um there's a difference between some of the health use cases and

between some of the health use cases and an IP protected regime now I know some

an IP protected regime now I know some privacy Advocates who might have

privacy Advocates who might have concerns about uh the health uh training

concerns about uh the health uh training data as well um but you know to level

data as well um but you know to level set um protection for creativity and

set um protection for creativity and patents in particular um that is not a

patents in particular um that is not a statute it exists in the Constitution of

statute it exists in the Constitution of the United States and so it is very very

the United States and so it is very very important to protect um our IP regime go

important to protect um our IP regime go ahead Mike great thank you um I'll start

ahead Mike great thank you um I'll start out just explain a little bit I I work

out just explain a little bit I I work for a company called Advance I've been

for a company called Advance I've been in the content industry as a lawyer for

in the content industry as a lawyer for a very long time um Advance is uh

a very long time um Advance is uh started out in the newspaper business

started out in the newspaper business over hundred years ago with the Staten

over hundred years ago with the Staten Island Advance emphasis on ads um but uh

Island Advance emphasis on ads um but uh that Chang you've heard of Our Brands we

that Chang you've heard of Our Brands we all Condon Nast um so the New Yorker

all Condon Nast um so the New Yorker Vanity Fair Vogue uh and a host of

Vanity Fair Vogue uh and a host of Publications you know in newspapers

Publications you know in newspapers around the country the American Business

around the country the American Business Journal so if there's a San Francisco

Journal so if there's a San Francisco Business Journal that's ours um so a

Business Journal that's ours um so a bunch of Publications in the business

bunch of Publications in the business we're in other businesses as well but

we're in other businesses as well but our roots are really there um and I do

our roots are really there um and I do think as we kind of talk about the

think as we kind of talk about the impact on journalism I do think is AI a

impact on journalism I do think is AI a good thing is is almost an irrelevant

good thing is is almost an irrelevant question AI is here deep machine

question AI is here deep machine learning has been around for quite some

learning has been around for quite some time um the large language models that

time um the large language models that began to be P get all the public

began to be P get all the public attention uh around 18 months ago um are

attention uh around 18 months ago um are really fascinating they're going to do a

really fascinating they're going to do a lot of great things

lot of great things but as we talked about regulation today

but as we talked about regulation today one of the things we said is let's see

one of the things we said is let's see where there are problems develop and

where there are problems develop and begin to address them and I think

begin to address them and I think journalism is an area where um we're

journalism is an area where um we're starting to see really big problems and

starting to see really big problems and problems that um kind of the current

problems that um kind of the current framework is having a hard time solving

framework is having a hard time solving um so to start out with journalism is

um so to start out with journalism is really expensive you know we put

really expensive you know we put reporters In Harm's Way all the time the

reporters In Harm's Way all the time the New Yorkers been reporting from Ukraine

New Yorkers been reporting from Ukraine the David Remick who and chief of the

the David Remick who and chief of the New Yorker went to Israel on I think it

New Yorker went to Israel on I think it was October 8th uh to begin reporting

was October 8th uh to begin reporting from there we uh we had a reporter of

from there we uh we had a reporter of one of our newspapers actually shot in

one of our newspapers actually shot in the face with a rubber bullet by a

the face with a rubber bullet by a police officer covering the George Floyd

police officer covering the George Floyd trials this is hard stuff and AI isn't

trials this is hard stuff and AI isn't going to do it right they're not going

going to do it right they're not going to go out there they're not going to

to go out there they're not going to interview people they're not going to be

interview people they're not going to be on the line they're not going to and and

on the line they're not going to and and the AIS are claiming no responsibility

the AIS are claiming no responsibility either when we do it we're responsible

either when we do it we're responsible if we defame somebody we're responsible

if we defame somebody we're responsible for making it right we have fact

for making it right we have fact checkers and the like um AI may help

checkers and the like um AI may help with some of those things it may help us

with some of those things it may help us gather information it may help us get

gather information it may help us get you know into public records with

you know into public records with ideation and first drafts but these um

ideation and first drafts but these um the business models that we are seeing

the business models that we are seeing are not about we've heard about people

are not about we've heard about people reading and is it like people reading

reading and is it like people reading documents is it like people learning is

documents is it like people learning is it like people learning a style of music

it like people learning a style of music and then they come up with new songs in

and then they come up with new songs in that style the issue is commercial

that style the issue is commercial Enterprises profiting off of this

Enterprises profiting off of this content without kind of breaking the

content without kind of breaking the social contract that's associated with

social contract that's associated with the way journalism Works journalism gets

the way journalism Works journalism gets paid for by subscriptions by advertising

paid for by subscriptions by advertising by data um by understanding the people

by data um by understanding the people who are accessing it and what we're

who are accessing it and what we're seeing today are uh exposing large

seeing today are uh exposing large amounts of content collected from our

amounts of content collected from our newspapers and the thousands of

newspapers and the thousands of newspapers around the country um without

newspapers around the country um without any of that right it shows up it's

any of that right it shows up it's either somewh mzed in a summary at the

either somewh mzed in a summary at the beginning of a search um it's uh it's

beginning of a search um it's uh it's quoted extensively um we've seen it as a

quoted extensively um we've seen it as a gradual thing before the latest chat

gradual thing before the latest chat Bots we saw you know in the beginning

Bots we saw you know in the beginning the cases were that you fair use was

the cases were that you fair use was enough of the content so that it could

enough of the content so that it could help you locate the content somewhere

help you locate the content somewhere else so someone would click on a image

else so someone would click on a image for example the perfect 10 case because

for example the perfect 10 case because they saw a little tiny box and the

they saw a little tiny box and the Court's reason nobody's going to use a

Court's reason nobody's going to use a little tiny box instead of the full

little tiny box instead of the full picture and they'll click and they'll

picture and they'll click and they'll get access to it and then the publisher

get access to it and then the publisher will will see that traffic um over time

will will see that traffic um over time you know we've seen the amount of

you know we've seen the amount of content that got used and exposed

content that got used and exposed gradually increased sort of like the

gradually increased sort of like the boiling frog right you know you put frog

boiling frog right you know you put frog in the boiling water if you heat it

in the boiling water if you heat it slowly enough nobody complains um but

slowly enough nobody complains um but what you started to see was that there

what you started to see was that there was a lot more content that would show

was a lot more content that would show up on a search results page or a

up on a search results page or a Facebook page and that content would

Facebook page and that content would then stay within the four walls and in

then stay within the four walls and in those four walls other companies would

those four walls other companies would serve ads against it they would direct

serve ads against it they would direct that traffic to other ones of their

that traffic to other ones of their experiences and the journalism um and

experiences and the journalism um and the Publishers would get less and less

the Publishers would get less and less of it in AI That's now gone on overdrive

of it in AI That's now gone on overdrive so you know you'll see a summary of news

so you know you'll see a summary of news articles that will appear right at the

articles that will appear right at the front we've seen uh sometimes sometimes

front we've seen uh sometimes sometimes it's it's phony you know we'll see

it's it's phony you know we'll see things attributed to our Publications

things attributed to our Publications that we never wrote uh sometimes we'll

that we never wrote uh sometimes we'll see really Oddities because as the AIS

see really Oddities because as the AIS hallucinate you'll see a recipe from

hallucinate you'll see a recipe from Bona petite that won't work uh because

Bona petite that won't work uh because they substituted some things you know

they substituted some things you know trying to be clever with the language

trying to be clever with the language I've been gluing cheese on pizza for GL

I've been gluing cheese on pizza for GL cheese on pizza

cheese on pizza exactly uh the AI won't understand

exactly uh the AI won't understand sarcasm which that was it was a

sarcasm which that was it was a sarcastic remark on Reddit and uh and

sarcastic remark on Reddit and uh and the AI took it as gospel um so you're

the AI took it as gospel um so you're starting to see this kind of problems

starting to see this kind of problems but you know at the end of the day what

but you know at the end of the day what competition Concepts have fundamentally

competition Concepts have fundamentally started with the idea that there is a

started with the idea that there is a property right and that those property

property right and that those property rights then result in exchange of value

rights then result in exchange of value and those exchange of values help

and those exchange of values help competitive markets function right and

competitive markets function right and so what you're seeing here is the denial

so what you're seeing here is the denial of the property right these companies

of the property right these companies are claiming using fair use or whatever

are claiming using fair use or whatever um that uh they have no obligation

um that uh they have no obligation whatever uh to compensate either for

whatever uh to compensate either for training or for output enforcement is

training or for output enforcement is very difficult because everybody's

very difficult because everybody's getting a different result there's a

getting a different result there's a random number generator built in to the

random number generator built in to the way uh natural language processing large

way uh natural language processing large language models generate their output

language models generate their output and so everybody's getting a different

and so everybody's getting a different answer so I may find you know in a in a

answer so I may find you know in a in a uh enforcement world you may find that

uh enforcement world you may find that copy and take it down or litigate with

copy and take it down or litigate with somebody that's very difficult to do

somebody that's very difficult to do unless the AI itself um is careful and

unless the AI itself um is careful and has rules built into it so that it's not

has rules built into it so that it's not generating it there's really no other

generating it there's really no other way to protect and so we do need new

way to protect and so we do need new rules um I am concerned about uh you

rules um I am concerned about uh you know the courts it's expensive to get

know the courts it's expensive to get there takes a long time cases get

there takes a long time cases get settled you know established principles

settled you know established principles judges are really poorly positioned to

judges are really poorly positioned to actually decide these things because

actually decide these things because we're talking about big policy questions

we're talking about big policy questions and that's not really where the

and that's not really where the Judiciary belongs um in 1976 when they

Judiciary belongs um in 1976 when they passed the Copyright Act um Congress

passed the Copyright Act um Congress debated the fair use rules for a long

debated the fair use rules for a long time and ultimately decided that they

time and ultimately decided that they wouldn't decide and they would just let

wouldn't decide and they would just let the case law evolve and gave us four

the case law evolve and gave us four principles um they didn't really tell

principles um they didn't really tell the judges much to do with it and judges

the judges much to do with it and judges are deciding based on just the facts of

are deciding based on just the facts of the particular case in front of them and

the particular case in front of them and not not really thinking about not really

not not really thinking about not really in a position to think about the larger

in a position to think about the larger impact um on what they're doing so um I

impact um on what they're doing so um I agree with David who was talking about

agree with David who was talking about unfair competition this is a species of

unfair competition this is a species of unfair competition there's actually a

unfair competition there's actually a case um it's not really very valid any

case um it's not really very valid any longer it was back when there was

longer it was back when there was federal law of um uh in contracts and

federal law of um uh in contracts and torts but there's a case that goes all

torts but there's a case that goes all the way back to the 19 uh 20s I think

the way back to the 19 uh 20s I think maybe late teens where um a the

maybe late teens where um a the Associated Press had reporters who were

Associated Press had reporters who were sending wires from the I think those

sending wires from the I think those were battles of World War I and um

were battles of World War I and um another company called the international

another company called the international new service would just take the AP wires

new service would just take the AP wires and sell them uh to customers and the

and sell them uh to customers and the court found the Supreme Court found that

court found the Supreme Court found that was unfair competition um you can't do

was unfair competition um you can't do that and so um that concept uh is really

that and so um that concept uh is really important now maybe that will get baked

important now maybe that will get baked in through the cases through fair use

in through the cases through fair use after all fair use starts with Fair it

after all fair use starts with Fair it talks about competition and the impact

talks about competition and the impact uh that a use has has on the market for

uh that a use has has on the market for Content um maybe Congress will step in

Content um maybe Congress will step in uh some of the states have tried but

uh some of the states have tried but there's the courts have been expanding

there's the courts have been expanding uh the concept of preemption under the

uh the concept of preemption under the copyright law and so have been limiting

things like um enforcement of terms of use and the like but one way or another

use and the like but one way or another if the courts don't begin to understand

if the courts don't begin to understand this property right then uh these

this property right then uh these businesses will not be in a position to

businesses will not be in a position to engage in their activi which I think

engage in their activi which I think it's not just about saving a business

it's not just about saving a business it's not about saving something archaic

it's not about saving something archaic I think journalism is something

I think journalism is something fundamental to our country um it's

fundamental to our country um it's fundamental to democracy it's

fundamental to democracy it's fundamental to how business markets work

fundamental to how business markets work whole host of things um and we've seen

whole host of things um and we've seen tremendous job losses in this industry

tremendous job losses in this industry you know the industry is one5 the size

you know the industry is one5 the size in terms of Revenue than it was just 20

in terms of Revenue than it was just 20 years ago um it is it is a real

years ago um it is it is a real existential threat that these um

existential threat that these um non-compensated non-permissive uses of

non-compensated non-permissive uses of our content uh will generate and and and

our content uh will generate and and and one last thing is terms of thinking more

one last thing is terms of thinking more precisely about your question about

precisely about your question about competition um in a world where people

competition um in a world where people have to negotiate for these rights um

have to negotiate for these rights um you'll expect to see a lot more

you'll expect to see a lot more competition among the people who are

competition among the people who are providing this information to Consumers

providing this information to Consumers right everybody has a could negotiate

right everybody has a could negotiate for a unique offering uh that one person

for a unique offering uh that one person may have the New York Times and another

may have the New York Times and another might have the Wall Street Journal

might have the Wall Street Journal another may have something else they may

another may have something else they may have different music offerings and the

have different music offerings and the like but when they've decided that it's

like but when they've decided that it's all free then everybody has all of it

all free then everybody has all of it and that form of competition disappears

and that form of competition disappears and so there's not just this uh property

and so there's not just this uh property rights issue but there's a set of

rights issue but there's a set of competition issues that creep in as

well Kathy go ahead uh so my question is actually for David um you're a musician

actually for David um you're a musician right um you teach students who are

right um you teach students who are interested in entering the music

interested in entering the music industry about music licensing which is

industry about music licensing which is fantastic um we've just heard uh from

fantastic um we've just heard uh from Michael about all the ongoing litigation

Michael about all the ongoing litigation or a lot of the ongoing litigation uh I

or a lot of the ongoing litigation uh I want to turn actually before I turn I

want to turn actually before I turn I want to footnote that so I want you to

want to footnote that so I want you to know that you have a friend in

know that you have a friend in government um so to the extent that

government um so to the extent that litigation is expensive uh one of the

litigation is expensive uh one of the roles that we play in government is

roles that we play in government is commenting in these cases and filing

commenting in these cases and filing briefs and so I want you know you do

briefs and so I want you know you do have a friend so please reach out if

have a friend so please reach out if there's anything that we can do in any

there's anything that we can do in any individual case you're welcome to email

individual case you're welcome to email us and we can potentially play a role to

us and we can potentially play a role to shape the law as it goes so with that

shape the law as it goes so with that with that caveat with that footnote um I

with that caveat with that footnote um I want to talk more about the

want to talk more about the practicalities so uh in terms of uh

practicalities so uh in terms of uh copyright

licensing how is that evolving when it comes to the music industry and is

comes to the music industry and is collective licensing a viable option and

collective licensing a viable option and I know this is going to be three

I know this is going to be three questions uh so uh should we be thinking

questions uh so uh should we be thinking about performance rights organizations

about performance rights organizations mod model for

mod model for licensing so performance rights

licensing so performance rights licensing I should explain for the

licensing I should explain for the audience very briefly that's like if

audience very briefly that's like if you're a bar or a restaurant and you

you're a bar or a restaurant and you play music in your bar or restaurant or

play music in your bar or restaurant or especially if like you're a little venue

especially if like you're a little venue that might have open mic nights and a DJ

that might have open mic nights and a DJ or something in there or something like

or something in there or something like that you know bar like that you got use

that you know bar like that you got use they usually have like a BMI as cap or a

they usually have like a BMI as cap or a seesack all three of those licenses

seesack all three of those licenses which means you there's three songwriter

which means you there's three songwriter organ izations for public performance

organ izations for public performance rights it means you have a license from

rights it means you have a license from all three of those you can play any song

all three of those you can play any song that you want in the bar without getting

that you want in the bar without getting permission right you effectively get

permission right you effectively get permission from all the songwriters by

permission from all the songwriters by getting the licenses from their from BMI

getting the licenses from their from BMI ASAP and seesack right so it works

ASAP and seesack right so it works pretty well when you have like you have

pretty well when you have like you have a whole bunch of

a whole bunch of lenses and you have a whole bunch of lur

lenses and you have a whole bunch of lur like a bunch of songwriters right and

like a bunch of songwriters right and they're sort of balanced in sort of

they're sort of balanced in sort of their economic power and the the blanket

their economic power and the the blanket licenses allow you to sort of so you

licenses allow you to sort of so you don't have like you know if you had a

don't have like you know if you had a hundred lenses and 100 Lors what is that

hundred lenses and 100 Lors what is that 10,000 licenses that would be required

10,000 licenses that would be required for each of that right so you get you

for each of that right so you get you get this market efficiency out of that

get this market efficiency out of that right also those are like sort of low

right also those are like sort of low value transactions each one of those

value transactions each one of those like if somebody plays a song in a bar I

like if somebody plays a song in a bar I mean I might get like a thousandth of a

mean I might get like a thousandth of a penny or something like that and if it's

penny or something like that and if it's played on the radio on a radio station I

played on the radio on a radio station I might get like 75 cents or something

might get like 75 cents or something like that but those are all transactions

like that but those are all transactions that are too small to stand on their own

that are too small to stand on their own I don't think that this case AI you're

I don't think that this case AI you're not going to have little mom and pop AI

not going to have little mom and pop AI shops you know you're not going to go

shops you know you're not going to go down to El Camino Real and there's a

down to El Camino Real and there's a little shop like Joe's AI or something

little shop like Joe's AI or something like that right you're can have really

like that right you're can have really big companies right doing AI right so

big companies right doing AI right so that doesn't really match where

that doesn't really match where Collective licensing has been successful

Collective licensing has been successful right because for instance I would say

right because for instance I would say the collective licenses hasn't been very

the collective licenses hasn't been very successful with companies like uh say

successful with companies like uh say Google or Spotify or Amazon and stuff

Google or Spotify or Amazon and stuff like that because what they do there's

like that because what they do there's usually a court that oversees the

usually a court that oversees the collective licensing

collective licensing they just keep challenging every

they just keep challenging every negotiation right and they just sort of

negotiation right and they just sort of run out the clock or just spend so much

run out the clock or just spend so much money on attorneys that they eventually

money on attorneys that they eventually get the deal that they want

get the deal that they want right so I would caution against that

right so I would caution against that now if songwriters say wanted to

now if songwriters say wanted to voluntarily get together and offer some

voluntarily get together and offer some licenses and stuff like that I don't

licenses and stuff like that I don't think that's necessarily bad but I

think that's necessarily bad but I wouldn't want to

wouldn't want to see um you know the way essentially BMI

see um you know the way essentially BMI and ASCAP are do their licensing through

and ASCAP are do their licensing through a

a doj uh consent decree right um I

doj uh consent decree right um I wouldn't want to see that level of

wouldn't want to see that level of Regulation inserted into it so it it has

Regulation inserted into it so it it has some benefits and I'm not maybe I

some benefits and I'm not maybe I haven't seen the model that works but

haven't seen the model that works but I'm cautious about that right now or

I'm cautious about that right now or more than cautious I'm skeptical that

more than cautious I'm skeptical that that's the proper application do you

that's the proper application do you want to add something to that yeah well

want to add something to that yeah well let me let me add to that

let me let me add to that um I think ASCAP actually started as a

um I think ASCAP actually started as a private organization it wasn't wasn't a

private organization it wasn't wasn't a consent decree for I think more than 50

consent decree for I think more than 50 years of ASCAP um and why it became a

years of ASCAP um and why it became a consent decree Just J person can you

consent decree Just J person can you explain why it became a consent decree

explain why it became a consent decree became a consent decree because ASCAP

became a consent decree because ASCAP got so large and essentially had such a

got so large and essentially had such a large market share of the music Market

large market share of the music Market that um it got to the point where no

that um it got to the point where no body could do business without an es

body could do business without an es also the radio stations boycotted them

also the radio stations boycotted them too well there's there was another set

too well there's there was another set of commercial boycott is also not C yes

of commercial boycott is also not C yes we can spend a lot of time on music

we can spend a lot of time on music licensing it's actually really but but

licensing it's actually really but but the point I wanted to make is um I think

the point I wanted to make is um I think the the point of the consent decrease is

the the point of the consent decrease is different from the point of collective

different from the point of collective licensing and I I'm a big supporter of

licensing and I I'm a big supporter of private Collective licensing as a

private Collective licensing as a solution here there's a couple reason

solution here there's a couple reason voluntary but there's a couple reasons

voluntary but there's a couple reasons why one is the um the the AI companies

why one is the um the the AI companies talk about how hard it is they even in

talk about how hard it is they even in motions to dismiss they've even said

motions to dismiss they've even said it's impossible to do licensing so

it's impossible to do licensing so therefore we should be allowed to use it

therefore we should be allowed to use it all for free um I don't believe that's

all for free um I don't believe that's true I believe there are private Market

true I believe there are private Market Solutions where uh people who have

Solutions where uh people who have rights that they want to license could

rights that they want to license could license them in an organization and

license them in an organization and there's huge efficiencies of having uh

there's huge efficiencies of having uh especially for smaller newspapers and

especially for smaller newspapers and middle siiz uh you know Publications to

middle siiz uh you know Publications to be able to uh go one place there's one

be able to uh go one place there's one license or series of licenses the

license or series of licenses the licenses um are uh efficiently had one

licenses um are uh efficiently had one of the things that ASCAP is great at is

of the things that ASCAP is great at is you're that little bar you just go in

you're that little bar you just go in the license is written there's no

the license is written there's no negotiation you don't need any lawyers

negotiation you don't need any lawyers you know what the rate is it's a

you know what the rate is it's a percentage of your Revenue so the small

percentage of your Revenue so the small guys pay less it's a very efficient

guys pay less it's a very efficient system that's built up um as a result of

system that's built up um as a result of the ne necessity so to me if you can get

the ne necessity so to me if you can get the property right correct then the

the property right correct then the private Market can handle the question

private Market can handle the question about Licensing in a efficient way and

about Licensing in a efficient way and so um I don't think that's that's really

so um I don't think that's that's really a response and I do think that there may

a response and I do think that there may be a lot more AI companies than you

be a lot more AI companies than you think today we're talking about

think today we're talking about Foundation models um large language

Foundation models um large language models things at the frontier um

models things at the frontier um tomorrow we may be talking about tens of

tomorrow we may be talking about tens of thousands of refined models and tens of

thousands of refined models and tens of thousands of businesses that are uh

thousands of businesses that are uh interested in using AI to help them

interested in using AI to help them surface journalism um I don't know what

surface journalism um I don't know what that all looks like but I think

that all looks like but I think something that's not government

something that's not government regulated is is a better way to do it

regulated is is a better way to do it now that doesn't mean that competition

now that doesn't mean that competition rules would have to be followed and the

rules would have to be followed and the normal you know kinds of things that

normal you know kinds of things that govern business practices would have to

govern business practices would have to be followed and it's possible that

be followed and it's possible that problems develop but where you would

problems develop but where you would start is a business that can adapt

start is a business that can adapt really fast nothing that's legislative

really fast nothing that's legislative adapts that quickly um that can deal

adapts that quickly um that can deal with things not just price but also what

with things not just price but also what the rules are and how it looks and what

the rules are and how it looks and what kinds of attribution you have to have

kinds of attribution you have to have and what kinds of links you have to

and what kinds of links you have to provide those kinds of things can be

provide those kinds of things can be flexible and negotiated uh between the

flexible and negotiated uh between the users and the and the copyright

users and the and the copyright community so I think that is a solution

community so I think that is a solution it only exists in the context of U of an

it only exists in the context of U of an enforcable property right hey

enforcable property right hey Michael it's

John from Los Angeles first of off I think you're naive to think there's

think you're naive to think there's going to be lots of AI

going to be lots of AI companies the uh Wall Street Journal

companies the uh Wall Street Journal said

said yesterday that um you know the average

yesterday that um you know the average AI chatbot weer needs up to 10 times as

AI chatbot weer needs up to 10 times as much electricity to process this an

much electricity to process this an internet

internet search and so there's a reason for

search and so there's a reason for instance that Microsoft's water use has

instance that Microsoft's water use has gone up by 45% in the last nine months

gone up by 45% in the last nine months it's and and there's a reason that open

it's and and there's a reason that open AI never succeeded until it had access

AI never succeeded until it had access to the Microsoft cloud as your platform

to the Microsoft cloud as your platform uh you and I have heard for many years

uh you and I have heard for many years like oh there's going to be competition

like oh there's going to be competition against Google in search never happened

against Google in search never happened uh there's going to be competition

uh there's going to be competition against Facebook in social media never

against Facebook in social media never happened um so the bigger going to get

happened um so the bigger going to get bigger and the thing about AI

bigger and the thing about AI is because you need so much data you

is because you need so much data you need so much computing power you need so

need so much computing power you need so much electricity you need so much money

much electricity you need so much money there these four companies that are

there these four companies that are dominating it now I promise you will be

dominating it now I promise you will be dominating it in 10

dominating it in 10 years I want to say that to me this

years I want to say that to me this question is only simple you have to

question is only simple you have to start by clarifying that nobody can

start by clarifying that nobody can scrape a copyrighted work off the

scrape a copyrighted work off the internet under the GU of fair use to

internet under the GU of fair use to train their models just full stop and

train their models just full stop and then you have to

then you have to say okay

say okay I agree that individual companies can

I agree that individual companies can negotiate with these big AI firms to

negotiate with these big AI firms to license their content for training it's

license their content for training it's already starting to happen I mean Wall

already starting to happen I mean Wall Street Journal has done it some other

Street Journal has done it some other companies have decided to do it so

companies have decided to do it so that's going to happen but then there

that's going to happen but then there has to be some kind of success feed for

has to be some kind of success feed for the individual artist if a million

the individual artist if a million people ask chat GPT write a Stephen King

people ask chat GPT write a Stephen King story about their Uncle Stephen King

story about their Uncle Stephen King should be getting something because Chad

should be getting something because Chad GPT learned how to write a Stephen King

GPT learned how to write a Stephen King short story by ingesting every Stephen

short story by ingesting every Stephen King short story that ever existed

King short story that ever existed without permission you know this whole

without permission you know this whole idea of permissionless innovation which

idea of permissionless innovation which these people live on that's why the

these people live on that's why the title of my first book was called move

title of my first book was called move fast and break things that's their

fast and break things that's their thesis we just going to do it and then

thesis we just going to do it and then then we'll try and ask for permission

then we'll try and ask for permission later and that's the way it is look

later and that's the way it is look these plutocrats that run these big AI

these plutocrats that run these big AI companies hear nothing about artists

companies hear nothing about artists except when they want to book Rihanna

except when they want to book Rihanna for their million dooll birthday

for their million dooll birthday [Laughter]

[Laughter] bars malib Mansion you know so I have

bars malib Mansion you know so I have one last thing I would ask everybody to

one last thing I would ask everybody to Google shrimp

Jesus okay so shrimp Jesus is an AI created image of Jesus as the head of a

created image of Jesus as the head of a shrimp there are 10,000 of those shrimp

shrimp there are 10,000 of those shrimp jesuses on

jesuses on Facebook they were generated by AI Bots

Facebook they were generated by AI Bots they were manipulated by AI Bots to make

they were manipulated by AI Bots to make money and at some point

money and at some point AI will take over the Internet

AI will take over the Internet completely and be doing all the

completely and be doing all the transactions um and you know the the

transactions um and you know the the agents the AI agents that are running

agents the AI agents that are running these shrimp creating the shrimp jesuses

these shrimp creating the shrimp jesuses and then putting them out they're

and then putting them out they're designed to farm engagement

designed to farm engagement clicks and they're making money on it I

clicks and they're making money on it I think and whatever the next shrimp Jesus

think and whatever the next shrimp Jesus thing is will all be created and there's

thing is will all be created and there's no humans involved whatsoever yeah I I

no humans involved whatsoever yeah I I look I think you identify a different

look I think you identify a different problem I don't think you're really

problem I don't think you're really arguing why Collective licensing doesn't

arguing why Collective licensing doesn't work but let me let me just add to one

work but let me let me just add to one thing which is AI can be part of the

thing which is AI can be part of the solution if people have the right

solution if people have the right incentives so for example there are AI

incentives so for example there are AI detectors out there that can tell when

detectors out there that can tell when something's written by AI um and they're

something's written by AI um and they're starting to be implemented um and they

starting to be implemented um and they use AI for that purpose to try to

use AI for that purpose to try to prevent for example uh you know AI

prevent for example uh you know AI submissions we we we're using some of

submissions we we we're using some of them to help prevent AI submissions to

them to help prevent AI submissions to the New Yorker for example in uh In

the New Yorker for example in uh In Articles um the uh um the uh I

Articles um the uh um the uh I understand Amazon for example uh AI

understand Amazon for example uh AI written books has become as you point

written books has become as you point out a real problem so there may be AI

out a real problem so there may be AI related Solutions some these if people

related Solutions some these if people have the right incentives uh in place to

have the right incentives uh in place to actually want to deal with those issues

actually want to deal with those issues well let's go back to the entertainment

well let's go back to the entertainment industry and are you concerned Duncan

industry and are you concerned Duncan about competition from shrimp Jesus I

about competition from shrimp Jesus I mean what's what's what are you doing at

mean what's what's what are you doing at sagaa uh when it comes to the use of AI

sagaa uh when it comes to the use of AI in the entertainment industry well

in the entertainment industry well before we move on because music is part

before we move on because music is part of what we do as well and I just want to

of what we do as well and I just want to say I mean we've talked a lot about the

say I mean we've talked a lot about the ASCAP you know BMI you know consent

ASCAP you know BMI you know consent decrease structure I think anybody who's

decrease structure I think anybody who's thinking about this if you're looking at

thinking about this if you're looking at the music industry with with all due

the music industry with with all due respect to the antitrust folks there are

respect to the antitrust folks there are other structures and the Sound Exchange

other structures and the Sound Exchange model is another example of that that

model is another example of that that didn't flow from a consent decree or

didn't flow from a consent decree or from an Anti-Trust scenario but flowed

from an Anti-Trust scenario but flowed from an actual you know section 114 of

from an actual you know section 114 of the Copyright Act a conscious decision

the Copyright Act a conscious decision to create a statutory licensing

to create a statutory licensing mechanism for non-interactive Digital

mechanism for non-interactive Digital streaming of music as it relates to

streaming of music as it relates to Performance royalties and uh you know

Performance royalties and uh you know I'm not I am not advocating for or

I'm not I am not advocating for or against a collective uh you know

against a collective uh you know management approach or a statutory

management approach or a statutory licensing approach in this area at this

licensing approach in this area at this point I think there's a lot that we need

point I think there's a lot that we need to figure out about what's going to

to figure out about what's going to happen on the fair use side but there

happen on the fair use side but there are ways to structure this I mean that

are ways to structure this I mean that that particular system system has proved

that particular system system has proved to be very beneficial for artists um in

to be very beneficial for artists um in part because you know when you think

part because you know when you think about this from the perspective of the

about this from the perspective of the user and the rights holder and then you

user and the rights holder and then you know Jonathan said something like you

know Jonathan said something like you know well and there needs to be a

know well and there needs to be a success payment for the artist there

success payment for the artist there needs to be that the artist part needs

needs to be that the artist part needs to be baked into the system in a way

to be baked into the system in a way that actually protects them and that's

that actually protects them and that's something that actually happened in The

something that actually happened in The Sound Exchange model so for example in

Sound Exchange model so for example in The Sound Exchange model the Record La

The Sound Exchange model the Record La labels share those proceeds with the

labels share those proceeds with the artist on a defined formula that's

artist on a defined formula that's statuto prescribed and that money is

statuto prescribed and that money is distributed directly to artists so there

distributed directly to artists so there can't be recruitment or offsets or the

can't be recruitment or offsets or the artist money does not get um applied to

artist money does not get um applied to advances or other things like that so I

advances or other things like that so I just I point it out not because I'm

just I point it out not because I'm saying it's the perfect model or that

saying it's the perfect model or that that necessarily statutory licensing is

that necessarily statutory licensing is the right approach but that there are a

the right approach but that there are a variety of approaches that can be taken

variety of approaches that can be taken in that space and anyone thinking about

in that space and anyone thinking about that from a policy perspective should

that from a policy perspective should look at the different options and not

look at the different options and not only focus on the option that's been out

only focus on the option that's been out there with songwriters and of course

there with songwriters and of course there's the mlc for publishing rights

there's the mlc for publishing rights the the music licens and Collective

the the music licens and Collective approach it's a newer part that came out

approach it's a newer part that came out of the music modernization act and some

of the music modernization act and some people would say that that you know

people would say that that you know remains to be seen how effective that is

remains to be seen how effective that is but those are those are options that

but those are those are options that need to be considered when uh but to to

need to be considered when uh but to to your question Kathy as we move back into

your question Kathy as we move back into you know the broader entertainment

you know the broader entertainment industry I would say that that a few

industry I would say that that a few things one I would say Collective

things one I would say Collective bargain you know has proved to be a very

bargain you know has proved to be a very effective tool in addressing AI so far

effective tool in addressing AI so far in the sense that it can move

all the workers in the industry at the conclusion of that period of time so

conclusion of that period of time so basically in less than a year there in

basically in less than a year there in our case we now have 16 pages of

our case we now have 16 pages of detailed contractual rules about the use

detailed contractual rules about the use of AI technology for digital replication

of AI technology for digital replication of actors including Voice work including

of actors including Voice work including generative AI provisions and since that

generative AI provisions and since that time we've negotiated Provisions with

time we've negotiated Provisions with the major record labels for the music

the major record labels for the music industry for television animation and

industry for television animation and for certain companies working in digital

for certain companies working in digital voice replication for video games

voice replication for video games although we're still fighting this

although we're still fighting this battle with some of the big video game

battle with some of the big video game companies and I say this only because in

companies and I say this only because in that same period of time uh public

that same period of time uh public policy really hasn't Advanced as much as

policy really hasn't Advanced as much as we needed to and in order to protect

we needed to and in order to protect people humans and make sure that AI

people humans and make sure that AI serves people and not the other way

serves people and not the other way around we need a mosaic of protection

around we need a mosaic of protection and part of that can be collective

and part of that can be collective bargaining and I think should be because

bargaining and I think should be because um it's it's been it's been able to keep

um it's it's been it's been able to keep pace a way that public policy hasn't but

pace a way that public policy hasn't but we also need public policy both

we also need public policy both regulatory legislative uh to to really

regulatory legislative uh to to really pick up the pace a bit and my pitch to

pick up the pace a bit and my pitch to everyone would be we cannot fall into

everyone would be we cannot fall into the Trap um that some especially tech

the Trap um that some especially tech companies want us to fall into of saying

companies want us to fall into of saying we need to come up with some

we need to come up with some comprehensive answer to all the problems

comprehensive answer to all the problems and issues of AI and bake that into some

and issues of AI and bake that into some comprehensive Omnibus legislation or

comprehensive Omnibus legislation or regulation and until that day comes

regulation and until that day comes nothing can be done like that can't be

nothing can be done like that can't be be the answer because people are

be the answer because people are suffering from these problems right now

suffering from these problems right now and I also would just say and I really

and I also would just say and I really appreciated your remarks during your fir

appreciated your remarks during your fir side chat earlier about the importance

side chat earlier about the importance of engaging internationally as well we

of engaging internationally as well we need to address this at the

need to address this at the international level and um as much as I

international level and um as much as I know a lot of governments aren't excited

know a lot of governments aren't excited about this that means we have to prepare

about this that means we have to prepare to move into a norm setting mode um I

to move into a norm setting mode um I was just at wio the world intellectual

was just at wio the world intellectual property organization a few months ago

property organization a few months ago or less than a few months ago a month

or less than a few months ago a month ago and a number of governments are

ago and a number of governments are saying you we want to do information

saying you we want to do information gathering we want to do more study we

gathering we want to do more study we want to do all this and I know the

want to do all this and I know the International System moves slowly but um

International System moves slowly but um we are at real risk of completely uh

we are at real risk of completely uh yielding the field here to tech

yielding the field here to tech companies that will just charge forward

companies that will just charge forward until some Norms are established and I

until some Norms are established and I really hope that the US government the

really hope that the US government the EU in particular as well as many others

EU in particular as well as many others will actually prepare themselves to move

will actually prepare themselves to move into a mindset of doing at least some

into a mindset of doing at least some Norm setting especially on image

Norm setting especially on image likeness I realize this is my pet issue

likeness I realize this is my pet issue but image likeness voice because it is

but image likeness voice because it is so personal and it is ready to go now

so personal and it is ready to go now like these are norms we can agree on and

like these are norms we can agree on and so uh I couldn't help but take the

so uh I couldn't help but take the opportunity Kathy while you're here to

opportunity Kathy while you're here to pitch that in your direction because I

pitch that in your direction because I know you have a lot to do with that so

know you have a lot to do with that so um but but I do think you know there is

um but but I do think you know there is we can make progress and this is not

we can make progress and this is not about saying no AI it's not about saying

about saying no AI it's not about saying block AI I know there are people there

block AI I know there are people there are members of mine who would like that

are members of mine who would like that to be our position but actually if you

to be our position but actually if you want to see AI my belief is if you want

want to see AI my belief is if you want to see AI um implemented in a way way

to see AI um implemented in a way way that is human- centered and respects

that is human- centered and respects people you can't just say no to it you

people you can't just say no to it you have to take the opportunity to channel

have to take the opportunity to channel it in the right direction and you kind

it in the right direction and you kind of give up that opportunity if you don't

of give up that opportunity if you don't engage on that point so I think all of

engage on that point so I think all of us have a responsibility to do that and

us have a responsibility to do that and my last point on this point is just to

my last point on this point is just to say we have to remember that humans are

say we have to remember that humans are implementing ai ai is not just

implementing ai ai is not just implementing itself it is not you know

implementing itself it is not you know people make the decisions that result in

people make the decisions that result in ai's implementation in every industry

ai's implementation in every industry government corporate everyone all of us

government corporate everyone all of us as people have a right to participate in

as people have a right to participate in that decision-making process and so I

that decision-making process and so I hope that everyone who cares about this

hope that everyone who cares about this which should be everyone uh step forward

which should be everyone uh step forward and make their voices heard because

and make their voices heard because otherwise you know it is just going to

otherwise you know it is just going to be a smaller group of people who are

be a smaller group of people who are making those decisions and the rest of

making those decisions and the rest of us will have to live with the outcome

us will have to live with the outcome well I say one thing about Collective

well I say one thing about Collective but because Collective licensing one

but because Collective licensing one thing it's just very small short thing

thing it's just very small short thing when I license a song for a bar to play

when I license a song for a bar to play they then don't go and create a

they then don't go and create a derivative work and use that to compete

derivative work and use that to compete against me right that's that's still

against me right that's that's still fundamentally an unfair competition

fundamentally an unfair competition issue yeah I want to add let me add one

issue yeah I want to add let me add one other real quick Point um which K off

other real quick Point um which K off some something that you just said um you

some something that you just said um you know actually AI Innovation and AI

know actually AI Innovation and AI progress actually requires that these

progress actually requires that these rights be protected and I'll explain why

rights be protected and I'll explain why today um for these natural language

today um for these natural language processes and image creation and all

processes and image creation and all these other things they're using all

these other things they're using all this human generated content and and um

this human generated content and and um if they don't have the human generated

if they don't have the human generated content they have nothing to train on

content they have nothing to train on and there have been studies that show

and there have been studies that show what happens when AI train on AI

what happens when AI train on AI generated content and the models break

generated content and the models break down and so it's a little like clear

down and so it's a little like clear cutting a forest you know to to make

cutting a forest you know to to make wood when the forest is gone you're

wood when the forest is gone you're stuck and so there are reasons why you

stuck and so there are reasons why you know for competition reasons why people

know for competition reasons why people don't want to pay they don't want to pay

don't want to pay they don't want to pay because their competitors aren't paying

because their competitors aren't paying they don't want to pay because they can

they don't want to pay because they can get away with it for now and they'll

get away with it for now and they'll deal with it later but eventually if

deal with it later but eventually if these industries and the work of these

these industries and the work of these artists is is deprecated then you not

artists is is deprecated then you not only damage those folks but you actually

only damage those folks but you actually damage the Machinery that's dependent on

damage the Machinery that's dependent on uh on this content to to

uh on this content to to exist I mean I I completely agree with

exist I mean I I completely agree with that statement and in fact it's possible

that statement and in fact it's possible that in certain collective bargaining

that in certain collective bargaining rooms I've said to the CEOs of these

rooms I've said to the CEOs of these companies big companies you know don't

companies big companies you know don't kill the goose that laid the golden egg

kill the goose that laid the golden egg your the reason why your company is

your the reason why your company is special is actually the relationship you

special is actually the relationship you have with creative talent and if you

have with creative talent and if you don't have that relationship there is no

don't have that relationship there is no reason why you are any different than a

reason why you are any different than a hundred other versions of your company

hundred other versions of your company that could go out there and simply just

that could go out there and simply just create mediocre content using existing

create mediocre content using existing algorithmic methodology to do that and

algorithmic methodology to do that and so um convincing the public that they

so um convincing the public that they like mediocre content isn't in your

like mediocre content isn't in your interest and damaging the very thing

interest and damaging the very thing that makes your company unique is also

that makes your company unique is also not in your own economic

not in your own economic interest so I think that's a fascinating

interest so I think that's a fascinating Point um I was recently listening to a

Point um I was recently listening to a conversation about um you know without

conversation about um you know without calling out a particular companies or

calling out a particular companies or people a um potential voice

people a um potential voice cloning thing that happened and um you

cloning thing that happened and um you know this Insight was really powerful um

know this Insight was really powerful um and it's it was that AI is still

and it's it was that AI is still relatively new right the public was just

relatively new right the public was just starting to um be curious about it be

starting to um be curious about it be willing to use it but these incidents

willing to use it but these incidents that break down Trust between companies

that break down Trust between companies and their workers or Talent um and the

and their workers or Talent um and the public are really damaging to companies

public are really damaging to companies that are looking to make a go in AI or

that are looking to make a go in AI or AI related Industries and that's um

AI related Industries and that's um that's a really powerful Point uh in my

that's a really powerful Point uh in my opinion um it makes people scared to

opinion um it makes people scared to want to work with them you know it makes

want to work with them you know it makes people you know it's one of the reasons

people you know it's one of the reasons why in a collective licensing system one

why in a collective licensing system one of the big objections to that is there

of the big objections to that is there are a lot of creative Talent who want to

are a lot of creative Talent who want to be able to just say no it's not just

be able to just say no it's not just that they want to be paid for it they

that they want to be paid for it they want to choose whether the content

want to choose whether the content they've created is used as training

they've created is used as training material for AI systems independent of

material for AI systems independent of the economic incentive because they have

the economic incentive because they have a personal philosophy about it or a

a personal philosophy about it or a strong belief about it and so that's

strong belief about it and so that's going to hamper you know any mandatory

going to hamper you know any mandatory Collective licensing mechanism or could

Collective licensing mechanism or could be a term of the collective

be a term of the collective licenses you you tributing

licenses you you tributing dcan higher moral value values to these

dcan higher moral value values to these people to which

people to which people the artists or the

people the artists or the corporations but did happened with the

corporations but did happened with the Scarlett Johansson incident you know

Scarlett Johansson incident you know they wanted to pretend like it was

they wanted to pretend like it was Scarlett

Scarlett Johansson and they got someone to be as

Johansson and they got someone to be as close to her as you possibly could and

close to her as you possibly could and they went out and asked Scarlet if if

they went out and asked Scarlet if if she would actually do it herself and she

she would actually do it herself and she says I'm not at all interested and the

says I'm not at all interested and the day before they released the voice they

day before they released the voice they went and ask her again please don't you

went and ask her again please don't you want to do this they knew what they were

want to do this they knew what they were doing they move fast and break things

doing they move fast and break things they don't

they don't care I before this panel was huge to be

care I before this panel was huge to be clear I'm not a I'm not a when I was

clear I'm not a I'm not a when I was refering the companies I was referring

refering the companies I was referring to were not the companies you're

to were not the companies you're referring to I was referring to the

referring to I was referring to the studios the streamers and the record

studios the streamers and the record labels who do have extensive

labels who do have extensive relationships with artists and frankly I

relationships with artists and frankly I have to say the the major record labels

have to say the the major record labels have been a very positive force on the

have been a very positive force on the legislative front in helping I mean if

legislative front in helping I mean if you look back at the the hearing we did

you look back at the the hearing we did in the Senate just a couple weeks ago on

in the Senate just a couple weeks ago on the no fakes act uh one of the CEOs of

the no fakes act uh one of the CEOs of one of the record labors Robert record

one of the record labors Robert record Lael was robt Kinsel from Warner Warner

Lael was robt Kinsel from Warner Warner music group was there testifying

music group was there testifying supporting that bill which would help

supporting that bill which would help provide individual rights in image uh

provide individual rights in image uh likeness voice Etc so I think there are

likeness voice Etc so I think there are you know some of those companies are

you know some of those companies are really recognizing and he's not the only

really recognizing and he's not the only one there have other other labels too I

one there have other other labels too I didn't mean to I didn't mean to accuse

didn't mean to I didn't mean to accuse the studios or the record companies I'm

the studios or the record companies I'm saying the big four AI companies are

saying the big four AI companies are from a different world and they don't

from a different world and they don't really care I mean the studios couldn't

really care I mean the studios couldn't exist if they had bad relationships with

exist if they had bad relationships with your artists right right no I just want

your artists right right no I just want to make sure that you knew that's who I

to make sure that you knew that's who I wasn't referring to when you when you in

wasn't referring to when you when you in your in your remark but but you know it

your in your remark but but you know it is it is important and I think your your

is it is important and I think your your your choice to reference the Scarlett

your choice to reference the Scarlett Johansson example I think is is

Johansson example I think is is instructive because you know she did

instructive because you know she did feel very strongly about it uh obviously

feel very strongly about it uh obviously and um and also it also is a is an

and um and also it also is a is an example of how with enough public

example of how with enough public support and enough public attention um

support and enough public attention um you know that type of uh pressure can

you know that type of uh pressure can have an impact uh you know it's not an

have an impact uh you know it's not an accident that sky was paused and you

accident that sky was paused and you know that's because because a very

know that's because because a very effective Savvy approach was chosen to

effective Savvy approach was chosen to how to deal with it in my opinion and um

how to deal with it in my opinion and um and that's how we all have to be we all

and that's how we all have to be we all have to be strategic in these moments um

have to be strategic in these moments um in order to to hold on to our rights so

in order to to hold on to our rights so um they are going to chase us off of the

um they are going to chase us off of the stage or the broom soon um but uh as

stage or the broom soon um but uh as co-host privilege I'm going to try to

co-host privilege I'm going to try to sneak in a lightning round 20 seconds

sneak in a lightning round 20 seconds for each of you what is and having heard

for each of you what is and having heard all of this discussion today what is one

all of this discussion today what is one thing

thing that you believe the antitrust agencies

that you believe the antitrust agencies who are law enforcers not Regulators can

who are law enforcers not Regulators can do to protect Creator

do to protect Creator rights Duncan you start we'll go to

rights Duncan you start we'll go to Jonathan and then down the line well to

Jonathan and then down the line well to the extent that you can I think I think

the extent that you can I think I think it using your law enforcement powers to

it using your law enforcement powers to make sure that there is genuine uh real

make sure that there is genuine uh real competition in this space is important

competition in this space is important and I think some of the other panelists

and I think some of the other panelists highlighted the exact reasons why but

highlighted the exact reasons why but real competition and making sure that

real competition and making sure that the power here doesn't get Consolidated

the power here doesn't get Consolidated too much these few companies will

too much these few companies will actually help put consumers and artists

actually help put consumers and artists on a More Level Playing Field so I hope

on a More Level Playing Field so I hope that could happen I launched a class

that could happen I launched a class action law Su against Spotify hang on

action law Su against Spotify hang on wait a minute my turn

wait a minute my turn Dave okay well this follows that better

Dave okay well this follows that better but thank you

but thank you John I I I just simple you need to

John I I I just simple you need to clarify that they cannot take this stuff

clarify that they cannot take this stuff under fair use that's the simplest thing

under fair use that's the simplest thing you can do that you cannot take

you can do that you cannot take copyrighted in stuff off the

copyrighted in stuff off the internet uh just because it's publicly

internet uh just because it's publicly available doesn't make it public you

available doesn't make it public you know that anyone can have it and and if

know that anyone can have it and and if you did do do that and remove section

you did do do that and remove section 230 immunity for these people you get

230 immunity for these people you get some progress

some progress of Doubt yeah for the avoidance of doubt

of Doubt yeah for the avoidance of doubt we do not have the authority to uh

we do not have the authority to uh remove section

remove section 230 I I hear you okay go ahead uh I I I

230 I I hear you okay go ahead uh I I I launched a successful class action

launched a successful class action lawsuit against Spotify over another

lawsuit against Spotify over another issue we won

issue we won but they don't these companies don't

but they don't these companies don't care um this isn't your department but

care um this isn't your department but um I I would like to see some criminal

um I I would like to see some criminal Mass copyright infringement it's it's a

Mass copyright infringement it's it's a criminal Mass copyright infingement the

criminal Mass copyright infingement the antitrust division jurisdiction over um

antitrust division jurisdiction over um antitrust crimes um and certain Title 18

antitrust crimes um and certain Title 18 FR um so I guess two things one is love

FR um so I guess two things one is love to see justi Department support for the

to see justi Department support for the unfer competition idea um as a way to to

unfer competition idea um as a way to to deal with the uh with these particular

deal with the uh with these particular issues um and I'd also like to see

issues um and I'd also like to see justice department support for the idea

justice department support for the idea that uh rights holders can collectively

that uh rights holders can collectively negotiate um because that that will

negotiate um because that that will require some cooperation particularly

require some cooperation particularly when we're dealing with large powerful

when we're dealing with large powerful organizations on the other side where

organizations on the other side where there's real power mismatches and

there's real power mismatches and sometimes dealing with monopsonies on

sometimes dealing with monopsonies on one side um and uh and the risk of

one side um and uh and the risk of antitrust problems on the side of Rights

antitrust problems on the side of Rights holders uh getting together in order to

holders uh getting together in order to negotiate where there power imbalances

negotiate where there power imbalances okay so that concludes uh the spirited

okay so that concludes uh the spirited panel um thank you to all of you uh for

panel um thank you to all of you uh for um uh letting letting us go over a

um uh letting letting us go over a little bit um and we will uh look

little bit um and we will uh look forward to the next conversation awesome

forward to the next conversation awesome thank you appreciate that

fun hopefully we have it Lively enough to

from hey [Music]

[Music] elhan hi s how are you I am so good how

very very briefly was

all right could every take their seats we're going to get started with our last

we're going to get started with our last panel of the day our regulation

panel of the day our regulation Spotlight

panel before we begin with our regulation Spotlight panel I just wanted

regulation Spotlight panel I just wanted to let everyone know that the antitrust

to let everyone know that the antitrust division is accepting comments on all of

division is accepting comments on all of the uh topics that we've discussed today

the uh topics that we've discussed today during our Workshop so if you have

during our Workshop so if you have something to say on any of those com uh

something to say on any of those com uh topics we would love to get your

topics we would love to get your thoughts and input and we will be

thoughts and input and we will be collecting those comments through July

collecting those comments through July 15th um and I'll give you our our uh our

15th um and I'll give you our our uh our email address uh box that you can submit

email address uh box that you can submit the comments to but it's also on our

the comments to but it's also on our website so you don't have to uh scourage

website so you don't have to uh scourage to write it all down now but uh ATR a

to write it all down now but uh ATR a 2024 AI workshop at us doj.gov

2024 AI workshop at us doj.gov uh so we'd love love to get your input

uh so we'd love love to get your input on some of the topics we've

on some of the topics we've discussed uh so I will uh turn it over

discussed uh so I will uh turn it over to David Lawrence who's going to be

to David Lawrence who's going to be moderating our regulation Spotlight uh

moderating our regulation Spotlight uh Dave is the policy director uh for the

Dave is the policy director uh for the antitrust Division and he's been in that

antitrust Division and he's been in that role since uh

role since uh 2021 and before that uh Dave has been in

2021 and before that uh Dave has been in a number of leadership roles in the

a number of leadership roles in the division uh so uh you are in good hands

division uh so uh you are in good hands with the regulation Spotlight here with

with the regulation Spotlight here with Dave moderating the panel and we will be

Dave moderating the panel and we will be discussing it uh during the panel uh you

discussing it uh during the panel uh you know I know today we've talked a lot

know I know today we've talked a lot about how and when to regulate AI uh so

about how and when to regulate AI uh so now we're going to talk about how

now we're going to talk about how regulation can support competition and

regulation can support competition and entry in the marketplace so I'll turn it

entry in the marketplace so I'll turn it over to Dave and just hang out for a

over to Dave and just hang out for a second Jen I want to acknowledge since

second Jen I want to acknowledge since we're coming up on the last panel Jen

we're coming up on the last panel Jen has not just been an MC today but she

has not just been an MC today but she was one of the tireless leaders of this

was one of the tireless leaders of this effort on the doj side along with a

effort on the doj side along with a great team at St so if we could just

great team at St so if we could just give her a round of

give her a round of applause thank you thank you appreciate

applause thank you thank you appreciate it uh and I'll also just congratulate

it uh and I'll also just congratulate and thank all of you who are still here

and thank all of you who are still here after eight hours of an incredibly I

after eight hours of an incredibly I think intellectually rigorous and

think intellectually rigorous and engaging event this has been so

engaging event this has been so substantive I have learned a lot we have

substantive I have learned a lot we have two treats as a reward for those of you

two treats as a reward for those of you who have stuck around at at least in

who have stuck around at at least in person only one for those

person only one for those online one of them is the reception that

online one of them is the reception that I saw them set out in the back uh so

I saw them set out in the back uh so you'll have to wait until after this

you'll have to wait until after this panel to get to that uh the second treat

panel to get to that uh the second treat though is I think a a panel that

though is I think a a panel that actually is going to help us shift from

actually is going to help us shift from what we've been learning today to how

what we've been learning today to how it's being implemented uh as regulators

it's being implemented uh as regulators and businesses are actively working to

and businesses are actively working to engage with the issues we've been

engage with the issues we've been discussing so I think it's a really nice

discussing so I think it's a really nice way to the end of the day thinking about

way to the end of the day thinking about this as Jen said it's it's sort of the

this as Jen said it's it's sort of the regulation panel and I think you under

regulation panel and I think you under the high top I

the high top I of uh regulation and competition in AI

of uh regulation and competition in AI there are really two like broad threads

there are really two like broad threads to set the table that we think about so

to set the table that we think about so one of them is that regulation can limit

one of them is that regulation can limit competition so even when regulation's

competition so even when regulation's necessary you could have regulations

necessary you could have regulations that are more limiting of competition

that are more limiting of competition than in others and there's sort of an

than in others and there's sort of an old proverb in DC that as long as there

old proverb in DC that as long as there have been monopolists there have been

have been monopolists there have been government Affairs offices who like the

government Affairs offices who like the kind of regulations that limit

kind of regulations that limit competition

competition right and so one of the things uh all of

right and so one of the things uh all of us need to be thinking about and that we

us need to be thinking about and that we be on the table here is what do

be on the table here is what do regulations that are very much needed to

regulations that are very much needed to address legitimate issues as they emerge

address legitimate issues as they emerge with AI what are the best ways to uh

with AI what are the best ways to uh apply and adapt those that maintain

apply and adapt those that maintain competition how should we be thinking

competition how should we be thinking about that and you know in what way

about that and you know in what way should we be sort of skeptical of

should we be sort of skeptical of approaches that are maybe a little bit

approaches that are maybe a little bit more heavy-handed uh sometimes designed

more heavy-handed uh sometimes designed and advocated by firms that have an

and advocated by firms that have an interest in keeping people out and then

interest in keeping people out and then the second threat is of course how do we

the second threat is of course how do we affirmatively promote competition

affirmatively promote competition through regulation and through law

through regulation and through law enforcement and that whether that be the

enforcement and that whether that be the antitrust laws Concepts like

antitrust laws Concepts like interoperability uh rules and

interoperability uh rules and regulations that enable the sort of Open

regulations that enable the sort of Open Source things we've talked about I mean

Source things we've talked about I mean all of those could be really

all of those could be really important uh and the reason this panel

important uh and the reason this panel is a treat is we have a tremendous I

is a treat is we have a tremendous I think group here to finish the day to go

think group here to finish the day to go over that I'm going to let you each

over that I'm going to let you each maybe introduce your interest in this a

maybe introduce your interest in this a bit but we have ellm tabasi who's a

bit but we have ellm tabasi who's a senior scientist at the National

senior scientist at the National Institute of Standards and technology

Institute of Standards and technology and the associate director for the

and the associate director for the emerging Technologies and the

emerging Technologies and the information technology laboratory she

information technology laboratory she leads n's trustworthy and responsible AI

leads n's trustworthy and responsible AI program so this is a real expert who's

program so this is a real expert who's been focused on these issues we have

been focused on these issues we have uter Desai who's the deputy chief

uter Desai who's the deputy chief technologist for Law and strategy at the

technologist for Law and strategy at the cfpb uh his work includes addressing the

cfpb uh his work includes addressing the implications for consumers

implications for consumers attributable to AI Technologies already

attributable to AI Technologies already out there and already interacting with

out there and already interacting with the legal regimes we have Professor

the legal regimes we have Professor Ellen Goodman from Ruckers who

Ellen Goodman from Ruckers who specializes in information policy law

specializes in information policy law and and I think thinks with a very uh

and and I think thinks with a very uh Broad and informed lens on these issues

Broad and informed lens on these issues and recently completed a stint uh as

and recently completed a stint uh as senior adviser for algorithmic Justice

senior adviser for algorithmic Justice at

at ntia uh and we have uh Dr Shanker panti

ntia uh and we have uh Dr Shanker panti who's at trust lab as a co-founder and

who's at trust lab as a co-founder and I'm really excited about this because

I'm really excited about this because it's it's spreading into a conversation

it's it's spreading into a conversation we've been having all day not a

we've been having all day not a government official but someone working

government official but someone working in

in Industry uh to help deal with the

Industry uh to help deal with the problems that regulation might address

problems that regulation might address but see how that can be done through

but see how that can be done through private sector um firms and support that

private sector um firms and support that where where the ecosystem will likely

where where the ecosystem will likely develop uh sort of ancillary Services I

develop uh sort of ancillary Services I think of this that help answer sort of

think of this that help answer sort of trust and safety questions so we'll just

trust and safety questions so we'll just start with a quick round Robin and if we

start with a quick round Robin and if we could just go down the line here and

could just go down the line here and then we'll hit you on the screen with

then we'll hit you on the screen with just a minute or two on your interest in

just a minute or two on your interest in these issues and kind of what you

these issues and kind of what you brought you to the AI

brought you to the AI conversation um sure so I think my

conversation um sure so I think my interest in AI just stems from its

interest in AI just stems from its growth in use in the markets for

growth in use in the markets for Consumer Financial goods and services it

Consumer Financial goods and services it is a place that we see really broadly

is a place that we see really broadly deployed you know whether it's in credit

deployed you know whether it's in credit underwriting in new uh fraud detection

underwriting in new uh fraud detection or really in customer service at the end

or really in customer service at the end of the the day there are a lot of

of the the day there are a lot of implications for consumers especially

implications for consumers especially when models don't work and I think a lot

when models don't work and I think a lot of the conversation we have is tends to

of the conversation we have is tends to be a little bit abstract but it's not so

be a little bit abstract but it's not so abstract when it's a consumer on the

abstract when it's a consumer on the other end who's interacting with a chat

other end who's interacting with a chat bot for example and can't get a straight

bot for example and can't get a straight answer or is stuck in a doom Loop so I

answer or is stuck in a doom Loop so I think a lot of my interest in the area

think a lot of my interest in the area comes from my experience um litigating

comes from my experience um litigating cases and uh and making sure that that

cases and uh and making sure that that consumers are put first great Ellen um

consumers are put first great Ellen um so uh I do all kinds of Technology law

so uh I do all kinds of Technology law and actually my background is in media

and actually my background is in media and Telecom law um I got into

and Telecom law um I got into algorithmic systems from the media side

algorithmic systems from the media side um sort of first touch with algorithmic

um sort of first touch with algorithmic systems which is social media and

systems which is social media and platforms um and became very interested

platforms um and became very interested in um sort of power and systemic risk

in um sort of power and systemic risk and also the interaction with the First

and also the interaction with the First Amendment um then I got into Ai and

Amendment um then I got into Ai and algorithmic systems um by through

algorithmic systems um by through studying local government use of um

studying local government use of um algorithmic algorithmic systems and that

algorithmic algorithmic systems and that made me interested again in a different

made me interested again in a different kinds of um Power relationships and lock

kinds of um Power relationships and lock in and maybe we'll we'll talk um a

in and maybe we'll we'll talk um a little bit about that um and then as

little bit about that um and then as David mentioned was um fortunate to

David mentioned was um fortunate to spend time in the government uh and uh

spend time in the government uh and uh the the out output of that um experience

the the out output of that um experience was a report on AI accountability policy

was a report on AI accountability policy and maybe we can talk a little bit about

and maybe we can talk a little bit about that today about um kind of the plumbing

that today about um kind of the plumbing of uh what of of AI accountability great

of uh what of of AI accountability great and Shanker yeah so I'll say a couple

and Shanker yeah so I'll say a couple sent sentences about trust lab so that

sent sentences about trust lab so that way people have uh some idea of where

way people have uh some idea of where I'm coming from so trust lab is a a

I'm coming from so trust lab is a a small startup in the Silicon Valley and

small startup in the Silicon Valley and we are in the business of helping online

we are in the business of helping online platforms uh like social media

platforms uh like social media marketplaces and so on keep that

marketplaces and so on keep that platform safe uh by that mean by that I

platform safe uh by that mean by that I mean keeping harmful content out uh so

mean keeping harmful content out uh so that whether it's hate speech harassment

that whether it's hate speech harassment misinformation disinformation terrorist

misinformation disinformation terrorist proper Anda things like that and our

proper Anda things like that and our recent Focus has been on automating

recent Focus has been on automating content moderation so a lot of the

content moderation so a lot of the platforms um have while they have

platforms um have while they have algorithms to moderate content the

algorithms to moderate content the policies tend to be pretty nuanced

policies tend to be pretty nuanced because they're trying to walk a fine

because they're trying to walk a fine line between free speech on one hand and

line between free speech on one hand and keeping platform safe on the other hand

keeping platform safe on the other hand and historically AI algorithms uh were

and historically AI algorithms uh were not good enough to moderate content uh

not good enough to moderate content uh but we are excited about the recent

but we are excited about the recent advances in artificial intelligence and

advances in artificial intelligence and how it can help automate uh some of the

how it can help automate uh some of the work that humans have to do today um

work that humans have to do today um which uh which is good because there are

which uh which is good because there are a lot of Wellness implications because

a lot of Wellness implications because for for for the low low paid uh jobs uh

for for for the low low paid uh jobs uh for the folks who are doing this this

for the folks who are doing this this work they have a lot of mental wellness

work they have a lot of mental wellness and other issues um but of course as we

and other issues um but of course as we automate this we also have to be mindful

automate this we also have to be mindful about the risk that we heard about today

about the risk that we heard about today including things like bias um so in

including things like bias um so in terms of relevance to the competition uh

terms of relevance to the competition uh we depend on hyperscalers a test lab

we depend on hyperscalers a test lab like many startups in the valley depends

like many startups in the valley depends on the

on the hyperscalers uh for for running our

hyperscalers uh for for running our business uh but we also compete with

business uh but we also compete with them because all of the hyperscalers and

them because all of the hyperscalers and and actually openi offer content

and actually openi offer content moderation services and apis so we we

moderation services and apis so we we have to both compete with them while

have to both compete with them while also depending on uh on them so that

also depending on uh on them so that that informs some of my perspectives

that informs some of my perspectives thanks and

thanks and Alum yes uh hello everyone sorry that I

Alum yes uh hello everyone sorry that I cannot be uh physically with all of you

cannot be uh physically with all of you and thank you for allowing me to join

and thank you for allowing me to join virtually I wish I had something very

virtually I wish I had something very insightful and good as other panelists

insightful and good as other panelists to say but really that the machine

to say but really that the machine learning a computer vision was my

learning a computer vision was my background training when I joined n that

background training when I joined n that my first project was development of a

my first project was development of a open vocabulary speaker recognition then

open vocabulary speaker recognition then with Biometrics I get more into um you

with Biometrics I get more into um you know developing models uh doing um

know developing models uh doing um neural networks and uh many other uh

neural networks and uh many other uh learning algorithms with particularly

learning algorithms with particularly with applications in Biometrics um and

with applications in Biometrics um and for a long time doing all of these

for a long time doing all of these things just because I could and I think

things just because I could and I think the question about should we and kind of

the question about should we and kind of responsible technology only hit me when

responsible technology only hit me when my children was were a kind of teenage

my children was were a kind of teenage high school and that was where the

high school and that was where the Facebook and all of these things were

Facebook and all of these things were just growing um and fast forward uh 2018

just growing um and fast forward uh 2018 I was uh 2017 2018 I was very lucky to

I was uh 2017 2018 I was very lucky to um have the opportunity to lead the

um have the opportunity to lead the trustworthy responsible AI program at n

trustworthy responsible AI program at n and think about uh all of these uh um

and think about uh all of these uh um opportunities uh but also challenges of

opportunities uh but also challenges of the technology in a uh deeper way and uh

the technology in a uh deeper way and uh and then uh several mandates and uh

and then uh several mandates and uh tasks that we got from Congress and

tasks that we got from Congress and executive orders uh gets us in the path

executive orders uh gets us in the path of development of the AI risk management

of development of the AI risk management uh engagement with the community and uh

uh engagement with the community and uh learning more about uh all of these uh

learning more about uh all of these uh uh opportunities and challenges that we

uh opportunities and challenges that we have with um emerging technology but in

have with um emerging technology but in this case particularly

this case particularly AI thank you so a panel that probably

AI thank you so a panel that probably like many in the audience came to these

like many in the audience came to these issues from very different perspectives

issues from very different perspectives but I I think those perspectives uh

but I I think those perspectives uh leveraging into this debate are really

leveraging into this debate are really important to get things right uh so I

important to get things right uh so I want to start actually Ellen with you

want to start actually Ellen with you and with the standard setting process

and with the standard setting process which you know nist is done a tremendous

which you know nist is done a tremendous amount of work here maybe just started

amount of work here maybe just started off with a a two-part question I think

off with a a two-part question I think of standard setting is a one of these

of standard setting is a one of these means for industry to in a way

means for industry to in a way self-regulate or to you know handle some

self-regulate or to you know handle some of the issues that may emerge with AI on

of the issues that may emerge with AI on its own um could you share a little bit

its own um could you share a little bit about the standard setting and AI work

about the standard setting and AI work that nist has been doing some of the

that nist has been doing some of the learnings and then second part maybe

learnings and then second part maybe answer that um issue I teased in the

answer that um issue I teased in the opener about the potential for the large

opener about the potential for the large maybe already powerful firms to try to

maybe already powerful firms to try to influence those processes to protect

influence those processes to protect their own power and and what we can do

their own power and and what we can do to make sure that small businesses

to make sure that small businesses medium businesses entrepreneurs have

medium businesses entrepreneurs have access to the

access to the process you're happy to uh thank you for

process you're happy to uh thank you for that question uh maybe I can just uh

that question uh maybe I can just uh first start by some clarifying that

first start by some clarifying that while standard is or our middle name is

while standard is or our middle name is part of our name um we develop voluntary

part of our name um we develop voluntary guidelines and framework to use by the

guidelines and framework to use by the community but we are not a standard

community but we are not a standard Setter organization the way uh I or uh

Setter organization the way uh I or uh ISO International standard organizations

ISO International standard organizations or and I think that's important

or and I think that's important distinctions to make and there are some

distinctions to make and there are some really good laws and policy in us that

really good laws and policy in us that standard setting is bottom up industry

standard setting is bottom up industry Le and the job of the federal government

Le and the job of the federal government is to um uh is to support convene and

is to um uh is to support convene and contribute to development of the

contribute to development of the standards rather than uh be the standard

standards rather than uh be the standard set um I'll talk a little bit more about

set um I'll talk a little bit more about all of these things but let me uh also

all of these things but let me uh also start by uh saying that as we heard

start by uh saying that as we heard today as you all know um AI is uh a

today as you all know um AI is uh a tremendous opportunity to improve our

tremendous opportunity to improve our lives but also negative consequences

lives but also negative consequences harms it's important to understand both

harms it's important to understand both and uh I can't tell it about ourselves

and uh I can't tell it about ourselves but I think it's correct for the almost

but I think it's correct for the almost everybody around the community that when

everybody around the community that when it comes to AI there is a lot Les we

it comes to AI there is a lot Les we know that we should or want to know U

know that we should or want to know U that's where uh

that's where uh um engaging in efforts to advance our

um engaging in efforts to advance our scientific understanding of the AI

scientific understanding of the AI models the behavior becomes important uh

models the behavior becomes important uh addressing ai's impact on people and

addressing ai's impact on people and society and Planet uh true Technical and

society and Planet uh true Technical and soci Technical lenses and solutions are

soci Technical lenses and solutions are important and advancing research on

important and advancing research on identifying and mitigating risks uh

identifying and mitigating risks uh including safety security privacy

including safety security privacy reliability in interoperability all of

reliability in interoperability all of these things becomes really important

these things becomes really important bottom line is that just like any other

bottom line is that just like any other technology when it comes to AI we want

technology when it comes to AI we want the technology that works works reliably

the technology that works works reliably accurately um a a it's a tech it's easy

accurately um a a it's a tech it's easy to do the right thing technology is

to do the right thing technology is developed in a way that's easy to do the

developed in a way that's easy to do the right thing and difficult to do the

right thing and difficult to do the wrong thing uh and if somebody

wrong thing uh and if somebody intentionally or unintentionally had

intentionally or unintentionally had done a wrong thing it's easy to recover

done a wrong thing it's easy to recover and bring it back to the right

and bring it back to the right directions that's where standards come

directions that's where standards come in into play um example of that is the

in into play um example of that is the three pin plug for the electricity that

three pin plug for the electricity that we use it's almost impossible to do to

we use it's almost impossible to do to use it in the wrong way uh that can

use it in the wrong way uh that can avoid a lot of the other accidents such

avoid a lot of the other accidents such as electric shocks and all this

as electric shocks and all this um uh for that reason standers play a

um uh for that reason standers play a really important role in development and

really important role in development and adoptions of the new technologies but

adoptions of the new technologies but for AI particularly becomes important

for AI particularly becomes important where as you said and many other

where as you said and many other panelists alluded to uh policy makers

panelists alluded to uh policy makers and Regulators in us around and abroad

and Regulators in us around and abroad you know globally are looking to the

you know globally are looking to the standard ecosystem uh to guide AI actors

standard ecosystem uh to guide AI actors on how to uh sort of implement uh high

on how to uh sort of implement uh high level principles and policies we have

level principles and policies we have seen it in some examples uh standards uh

seen it in some examples uh standards uh uh stay at the high level abstract level

uh stay at the high level abstract level of saying that AI systems must be safe

of saying that AI systems must be safe or secure or non-discriminatory but

or secure or non-discriminatory but don't get to the level of explaining and

don't get to the level of explaining and hashing out what we mean by

hashing out what we mean by non-discriminatory what we mean by

non-discriminatory what we mean by secure with safe and more importantly

secure with safe and more importantly what are the ways to test and ensure

what are the ways to test and ensure that the systems is

that the systems is nondiscriminatory safe or secures so

nondiscriminatory safe or secures so establishing terminology taxonomies uh

establishing terminology taxonomies uh specifications for trustworthiness

specifications for trustworthiness characteristics uh testing methodology

characteristics uh testing methodology confirmity assessments all of these

confirmity assessments all of these things becomes uh important activities

things becomes uh important activities that standards can uh build that uh uh

that standards can uh build that uh uh Foundation or sort of support for uh

Foundation or sort of support for uh policies and regulations regardless of

policies and regulations regardless of the um the landscape of what policy and

the um the landscape of what policy and what regulations um Can can provide some

what regulations um Can can provide some sort of a u

sort of a u a support for the implementations and

a support for the implementations and enforcements of of the laws uh but also

enforcements of of the laws uh but also provide and create a Level Playing Field

provide and create a Level Playing Field for uh

for uh Innovations um recently as part of a

Innovations um recently as part of a task that uh as part of the executive

task that uh as part of the executive order on safe secure and trustworthy AI

order on safe secure and trustworthy AI that was released on

that was released on October ask us to do several things

October ask us to do several things including uh developing a plan for

including uh developing a plan for Global engagement in development of AI

Global engagement in development of AI standards uh the draft is out for public

standards uh the draft is out for public comment by Sunday midnight I believe u

comment by Sunday midnight I believe u but I want to address your second part

but I want to address your second part of your questions uh that plan talks

of your questions uh that plan talks about standards AI standards ai ai

about standards AI standards ai ai standards landscape and then gets into

standards landscape and then gets into some of the activities and some of the

some of the activities and some of the recommendations for um Global engagement

recommendations for um Global engagement for development of the better standards

for development of the better standards um one one thing that we talk about uh

um one one thing that we talk about uh in that uh in that document uh we

in that uh in that document uh we address activities before during and

address activities before during and after creation of Standards um standard

after creation of Standards um standard setting usually are being discussed or

setting usually are being discussed or uh considered as going to the standard

uh considered as going to the standard meeting and developing the document but

meeting and developing the document but good standards really need a scientific

good standards really need a scientific Foundation foundations a scientific

Foundation foundations a scientific underpinning for those so there is there

underpinning for those so there is there is a need for research a foundation body

is a need for research a foundation body of scientific and Technical work uh uh

of scientific and Technical work uh uh before a standard being developed so

before a standard being developed so that's one type of Engagement and making

that's one type of Engagement and making sure that a broader Community are

sure that a broader Community are involved on building those technical

involved on building those technical building blocks that's important for the

building blocks that's important for the standards uh talking about the during uh

standards uh talking about the during uh standard development um many of the uh

standard development um many of the uh standard development processes has a due

standard development processes has a due process that

process that um um encourage or allows for inclusive

um um encourage or allows for inclusive open transparent processes and those are

open transparent processes and those are the standard development organizations

the standard development organizations that um we at least believe that they

that um we at least believe that they Merit participations but even after

Merit participations but even after standards are finalized developed um

standards are finalized developed um there is Need for engagements and

there is Need for engagements and scientific works on developing tools um

scientific works on developing tools um data sets benchmarks whatever is needed

data sets benchmarks whatever is needed to help with implementations and

to help with implementations and operationalizations of the standards

operationalizations of the standards again that becomes another area that

again that becomes another area that different voices different perspectives

different voices different perspectives different um uh backgrounds uh need to

different um uh backgrounds uh need to participate as part of the standard

participate as part of the standard again it talks about the importance of a

again it talks about the importance of a inclusive participations and um um how

inclusive participations and um um how we the things that we ought to do and we

we the things that we ought to do and we don't have the answer to all of those

don't have the answer to all of those things but at least being aware that

things but at least being aware that it's really important to have um to

it's really important to have um to include the different perspectives

include the different perspectives background and expertise in in um the

background and expertise in in um the research that's been done as part of the

research that's been done as part of the pre- standardization work standard

pre- standardization work standard setting and Post-Standard

setting and Post-Standard activities um it's really becomes

activities um it's really becomes important because we want to reflect the

important because we want to reflect the needs and input from a diverse Global

needs and input from a diverse Global set of uh stakeholders um U so that the

set of uh stakeholders um U so that the standards are context sensitive

standards are context sensitive performance-based human centered and we

performance-based human centered and we can only achieve that if we reach out to

can only achieve that if we reach out to a broader community on getting those uh

a broader community on getting those uh inputs uh and then the the process

inputs uh and then the the process should be open inclusive and transparent

should be open inclusive and transparent driven by consensus uh uh to make sure

driven by consensus uh uh to make sure that not only the input is sought but

that not only the input is sought but also included in the process um I don't

also included in the process um I don't know I I I I don't want to go over time

know I I I I don't want to go over time but uh I'll stop here and say that while

but uh I'll stop here and say that while uh it may sounds like that um and it can

uh it may sounds like that um and it can very much be true that in standard

very much be true that in standard development processes can be influenced

development processes can be influenced by certain participants um but because

by certain participants um but because of the due process because of the um uh

of the due process because of the um uh importance of getting the right

importance of getting the right technical data into the standard uh uh

technical data into the standard uh uh development discussions uh there are uh

development discussions uh there are uh mechanisms and opportunities to make

mechanisms and opportunities to make sure that uh we bring uh uh

sure that uh we bring uh uh participations and voices of a broader

participations and voices of a broader Community great thank you uh so I guess

Community great thank you uh so I guess in terms of bringing in the broad

in terms of bringing in the broad participation people should be rushing

participation people should be rushing out the door to prep comments for that

out the door to prep comments for that Sunday deadline you mentioned uh any

Sunday deadline you mentioned uh any comments from the on standards before we

comments from the on standards before we turn to the next

turn to the next topic yeah I'd I'd love to make a few um

topic yeah I'd I'd love to make a few um it was interesting that Dr Rice

it was interesting that Dr Rice mentioned um standard setting process

mentioned um standard setting process and processes and she talked about um

and processes and she talked about um how important it was who was in the room

how important it was who was in the room and how um sometimes uh the US had

and how um sometimes uh the US had missed the boat on in terms of um

missed the boat on in terms of um Staffing some of the technical

Staffing some of the technical committees um the other people who are

committees um the other people who are often not in the room are um smmes as as

often not in the room are um smmes as as you mentioned and also Civil Society um

you mentioned and also Civil Society um folks now you know formally they are Inc

folks now you know formally they are Inc these are inclusive processes and they

these are inclusive processes and they are encouraged to be in the room and

are encouraged to be in the room and there are all kinds of um uh both

there are all kinds of um uh both International and and domestic impetus

International and and domestic impetus um to have them in the room but the fact

um to have them in the room but the fact is it's very expensive and so you know

is it's very expensive and so you know we heard from a lot of Civil Society

we heard from a lot of Civil Society participants that said they just don't

participants that said they just don't have um the resources uh you know to to

have um the resources uh you know to to be there as consistently as you need to

be there as consistently as you need to um to have sway so that's um you know a

um to have sway so that's um you know a real um a real obstacle and just if I

real um a real obstacle and just if I can just make two other points about

can just make two other points about standards that maybe we can come back to

standards that maybe we can come back to um is I can't underscore enough elham's

um is I can't underscore enough elham's point about the importance of standard

point about the importance of standard setting and um the way in which

setting and um the way in which standards will be sort of incorporated

standards will be sort of incorporated into law either um formally or

into law either um formally or informally so you already see in the EU

informally so you already see in the EU AI act that you know they have a list of

AI act that you know they have a list of of standards that have to be um that

of standards that have to be um that they are looking towards and they are

they are looking towards and they are regulating to those standards um you

regulating to those standards um you know things like accuracy and robustness

know things like accuracy and robustness um and uh and I also think um we're

um and uh and I also think um we're going to begin to see in in litigation

going to begin to see in in litigation that you know the standards kind of set

that you know the standards kind of set the standard for um you know negligence

the standard for um you know negligence or safe harbors um burdens of proof and

or safe harbors um burdens of proof and so you know they end up becoming um even

so you know they end up becoming um even if they're voluntary uh industry

if they're voluntary uh industry standard standards they end up becoming

standard standards they end up becoming uh having um a duy uh impact and then

uh having um a duy uh impact and then one other point um that I just think is

one other point um that I just think is very difficult and I just want to flag

very difficult and I just want to flag is that um you know we have we can have

is that um you know we have we can have a standard for an electrical outlet um

a standard for an electrical outlet um which is a difficult and important

which is a difficult and important technical standard to set there aren't a

technical standard to set there aren't a lot of norms around what that should

lot of norms around what that should look like um there are a lot of norms

look like um there are a lot of norms around what discrimination looks like um

around what discrimination looks like um and what the standard should be for

and what the standard should be for fairness uh or for sustainabil ility are

fairness uh or for sustainabil ility are some of the things that we're talking

some of the things that we're talking about with AI so it adds a whole another

about with AI so it adds a whole another layer of complexity um to what we're

layer of complexity um to what we're asking uh standard setting organizations

asking uh standard setting organizations uh to do great and maybe let's shift a

uh to do great and maybe let's shift a little bit but stick with you for a

little bit but stick with you for a moment Ellen on a different question of

moment Ellen on a different question of layers of complexity uh I was thinking

layers of complexity uh I was thinking ahead to your discussion when we heard

ahead to your discussion when we heard from Andrew Ang this morning and he had

from Andrew Ang this morning and he had a wonderful discussion but he really

a wonderful discussion but he really advocated uh don't regulate technology

advocated uh don't regulate technology only regulate the application and so he

only regulate the application and so he had a an electric motor was a technology

had a an electric motor was a technology and then you know the blender and the uh

and then you know the blender and the uh the blender and the missile were the

the blender and the missile were the applications I was wishing he had an

applications I was wishing he had an electric bike down there because I

electric bike down there because I recently purchased one and I learned

recently purchased one and I learned that uh that they do regulate the motor

that uh that they do regulate the motor uh if it's more than 500 watts it's

uh if it's more than 500 watts it's subject to a different regulatory regime

subject to a different regulatory regime and in fact I think you look out in that

and in fact I think you look out in that parking lot you might know that if you

parking lot you might know that if you have a many horsepower electric motor

have a many horsepower electric motor there's various Federal authorities that

there's various Federal authorities that need to approve the use of that motor in

need to approve the use of that motor in the car before it's it's put in uh and

the car before it's it's put in uh and so it it did occur to me that sometimes

so it it did occur to me that sometimes there are measures of Power by which

there are measures of Power by which we're regulating at the technology layer

we're regulating at the technology layer and I know you have thought a lot about

and I know you have thought a lot about this sort of layers of Regulation

this sort of layers of Regulation question so I'd love uh for you to share

question so I'd love uh for you to share your thoughts with the audience here

your thoughts with the audience here yeah so I mean what's so interesting

yeah so I mean what's so interesting about that discussion is what is the

about that discussion is what is the right analogy and are you know we're

right analogy and are you know we're just taking llms you know it's it's it's

just taking llms you know it's it's it's infrastructure so is it like water and

infrastructure so is it like water and electricity um or is it you know

electricity um or is it you know something quite different from that I

something quite different from that I think it's quite different from that

think it's quite different from that because it is a dynamic system that's

because it is a dynamic system that's trained on um you know data that that

trained on um you know data that that has privacy components and and copyright

has privacy components and and copyright components we just heard heard about

components we just heard heard about there are certain um uh there's opacity

there are certain um uh there's opacity around these this infrastructure that is

around these this infrastructure that is not true of of an engine and I think

not true of of an engine and I think David to your to your point um you know

David to your to your point um you know with respect to engines or we could we

with respect to engines or we could we might say the same thing about you know

might say the same thing about you know chemicals or or Pharmaceuticals is is

chemicals or or Pharmaceuticals is is you know we have a regulatory structure

you know we have a regulatory structure which has both X Ane components so

which has both X Ane components so sometimes we don't regulate things um uh

sometimes we don't regulate things um uh ex Ane because we've got a vigorous

ex Ane because we've got a vigorous expost liability structure around it or

expost liability structure around it or sometimes um the opposite is true and so

sometimes um the opposite is true and so all of those features I think are

all of those features I think are important in answering the question you

important in answering the question you know do you and I I think to to put a

know do you and I I think to to put a fine point on what he was saying is it's

fine point on what he was saying is it's not the developers responsibility it's

not the developers responsibility it's the deployer responsibility and I think

the deployer responsibility and I think we can talk about whose responsibility

we can talk about whose responsibility it should be and who gets regulated

it should be and who gets regulated depending on the context sure sure

depending on the context sure sure others thoughts on this question

others thoughts on this question developers deployers what's the best

developers deployers what's the best place to regulate yeah sure I have some

place to regulate yeah sure I have some thoughts Dave um you know from my

thoughts Dave um you know from my perspective and this really comes from

perspective and this really comes from my my days as a litigator and our

my my days as a litigator and our enforcement division but I think you

enforcement division but I think you know question surrounding who to

know question surrounding who to determine as the responsible party is

determine as the responsible party is really a fact intensive inquiry you know

really a fact intensive inquiry you know one could hold responsible the developer

one could hold responsible the developer one could hold responsible the user of

one could hold responsible the user of the AI model one could hold both

the AI model one could hold both responsible so I know you know it's

responsible so I know you know it's everywh Lori's favorite question or uh

everywh Lori's favorite question or uh response but it depends and it really

response but it depends and it really does depend on the facts and the

does depend on the facts and the circumstances and I don't really want to

circumstances and I don't really want to get into a discussion of jurisdictions

get into a discussion of jurisdictions or things of that nature nature but I

or things of that nature nature but I want what I do want to make sure is

want what I do want to make sure is clear is that you know as a first

clear is that you know as a first principle we need to ensure that each

principle we need to ensure that each and every actor in for example the AI

and every actor in for example the AI ecosystem is adhering to their legal

ecosystem is adhering to their legal obligations and we don't leave consumers

obligations and we don't leave consumers holding the bag I think we know from the

holding the bag I think we know from the past that the one thing that does not

past that the one thing that does not work is a pass the buck type of

work is a pass the buck type of accountability system so

accountability system so yeah and so yeah jump in

yeah and so yeah jump in uh I also see a role for standards here

uh I also see a role for standards here um I agree that it should be a shared

um I agree that it should be a shared responsibility and accountability across

responsibility and accountability across the board you know it it may sound in a

the board you know it it may sound in a in a conceptual way um simple and and

in a conceptual way um simple and and non messy if you can just say um

non messy if you can just say um responsibility is with you know the end

responsibility is with you know the end of the chain but this doesn't really

of the chain but this doesn't really work in reality for a lot of reasons

work in reality for a lot of reasons including that you know if we um some

including that you know if we um some big industry is difficult to get them to

big industry is difficult to get them to to

to um yeah the level of accountability will

um yeah the level of accountability will be different but standards can play a

be different but standards can play a role here too and that gets into the uh

role here too and that gets into the uh transparency requirement and standards

transparency requirement and standards were reporting on what they had done

were reporting on what they had done standards reporting of the testing that

standards reporting of the testing that they had done so at any um any AI actors

they had done so at any um any AI actors across the value chain across this

across the value chain across this spectrum of the de designer developer

spectrum of the de designer developer fine tuners deployers uh if they if

fine tuners deployers uh if they if there is standards for uh transparency

there is standards for uh transparency on um actions being taken tests being

on um actions being taken tests being done and standards way of reporting that

done and standards way of reporting that that can help thank you I think as a

that can help thank you I think as a practitioner one one issue that I've uh

practitioner one one issue that I've uh I've seen is that when you're building

I've seen is that when you're building on top of foundational

on top of foundational models um it's it's different from some

models um it's it's different from some other type of technology in that it's

other type of technology in that it's not really well specified so it's not

not really well specified so it's not like for example databases many people

like for example databases many people build on top of databases but there is a

build on top of databases but there is a kind of SQL language which which has has

kind of SQL language which which has has a certain specification and then there

a certain specification and then there is the understanding that all databases

is the understanding that all databases are going to support SQL language uh but

are going to support SQL language uh but the large language models it it's bit of

the large language models it it's bit of a guessing game right mean what it's

a guessing game right mean what it's going to answer correctly and what it's

going to answer correctly and what it's not going to answer correctly and so

not going to answer correctly and so while on one hand we can say maybe it's

while on one hand we can say maybe it's the responsibility of the deployer to

the responsibility of the deployer to make sure that the llm is functioning

make sure that the llm is functioning correctly before they use it um but

correctly before they use it um but where it can become complicated is that

where it can become complicated is that it's not just building that initially

it's not just building that initially but ongoing maintenance so in the future

but ongoing maintenance so in the future when they have to upgrade uh let's say

when they have to upgrade uh let's say the new model did not behave the same

the new model did not behave the same way as the old model uh you're kind of

way as the old model uh you're kind of stuck in a bad place right where you can

stuck in a bad place right where you can either choose to not upgrade which is

either choose to not upgrade which is not a tenable solution uh or upgrade but

not a tenable solution uh or upgrade but have functionality change in ways that

have functionality change in ways that you can't control right so so in that

you can't control right so so in that sense I think this problem is a bit more

sense I think this problem is a bit more complicated just because of the uh lack

complicated just because of the uh lack of well specified nature of these

of well specified nature of these Foundation models which which which

Foundation models which which which which is also what makes them great

which is also what makes them great because they're kind of general purpose

because they're kind of general purpose technologies that can be used for many

technologies that can be used for many things but they do bring this challenge

things but they do bring this challenge I see so maybe let's let's let's take

I see so maybe let's let's let's take that premise we're we're working out is

that premise we're we're working out is it the developer layer the deployer

it the developer layer the deployer layer fact specific legal regime

layer fact specific legal regime specific uter uh maybe take us down the

specific uter uh maybe take us down the regulatory stack

regulatory stack here in the AI ecosystem like how do you

here in the AI ecosystem like how do you or how does the cfpb think about

or how does the cfpb think about remedies in law enforcement what's

remedies in law enforcement what's what's different what's the same um yeah

what's different what's the same um yeah sure so

sure so at a very high level you know vigorous

at a very high level you know vigorous enforcement of pre-existing laws and

enforcement of pre-existing laws and thoughtful remedy design are two really

thoughtful remedy design are two really important components to ensuring that

important components to ensuring that the markets for Consumer Financial goods

the markets for Consumer Financial goods and services remain competitive and to

and services remain competitive and to helping to ensure that a firm's breaking

helping to ensure that a firm's breaking the law does not confer upon it a

the law does not confer upon it a competitive Advantage so let me take a

competitive Advantage so let me take a minute and unpack that a little bit and

minute and unpack that a little bit and talk about the cfpb's efforts in this

talk about the cfpb's efforts in this area so first in thinking about the

area so first in thinking about the rules and the guidance themselves one

rules and the guidance themselves one point one point that the cfpb has made

point one point that the cfpb has made clear is that markets work best when

clear is that markets work best when rules are simple easy to understand and

rules are simple easy to understand and easy to enforce in other words when

easy to enforce in other words when there are bright lines so in the context

there are bright lines so in the context of AI for example the cfpb has stated a

of AI for example the cfpb has stated a number of times that there is no AI or

number of times that there is no AI or complex technology exception to Federal

complex technology exception to Federal Consumer Financial laws and making this

Consumer Financial laws and making this a little bit more tangible the cfbb has

a little bit more tangible the cfbb has offered Guidance with regard to Ai and

offered Guidance with regard to Ai and blackbox models and their use in credit

blackbox models and their use in credit decision

decision you know when talking about AI we hear a

you know when talking about AI we hear a lot of discussions about model

lot of discussions about model explainability what does it mean what

explainability what does it mean what are the obligations you know similar

are the obligations you know similar questions of that nature but Federal

questions of that nature but Federal Consumer Financial law has actually long

Consumer Financial law has actually long had an explainability requirement

had an explainability requirement specifically the requirements under the

specifically the requirements under the Equal Credit Opportunity Act or a COA

Equal Credit Opportunity Act or a COA you know that firm's making credit

you know that firm's making credit decisions must provide a notice to

decisions must provide a notice to Consumers if they make an adverse credit

Consumers if they make an adverse credit decision such as denying them cred

decision such as denying them cred credit so recently the cfpb issued two

credit so recently the cfpb issued two uh circulars related to this adverse

uh circulars related to this adverse action noes requirement without getting

action noes requirement without getting into the details of those circulars you

into the details of those circulars you know they we made clear there that

know they we made clear there that companies relying on complex algorithms

companies relying on complex algorithms must provide uh specific explanations

must provide uh specific explanations for denying credit accurate and specific

for denying credit accurate and specific um sorry provide accurate and specific

um sorry provide accurate and specific explanations for denying credit um

explanations for denying credit um applications and companies do not

applications and companies do not absolve their of their legal obligations

absolve their of their legal obligations simply because they let a blackbox model

simply because they let a blackbox model make the the decision a model where they

make the the decision a model where they might not actually understand why it's

might not actually understand why it's doing what it's doing so ultimately

doing what it's doing so ultimately again getting back to Bright line rules

again getting back to Bright line rules you know cfpb's approach is if using a

you know cfpb's approach is if using a complex model means that a company

complex model means that a company cannot comply with its obligations under

cannot comply with its obligations under Federal Consumer Financial laws then

Federal Consumer Financial laws then company should not be using that model

company should not be using that model and moving on a little bit to

and moving on a little bit to remedies you know when you have a

remedies you know when you have a violation of Law and are contemplating

violation of Law and are contemplating what remedies are appropriate

what remedies are appropriate it's important to make sure that

it's important to make sure that violating the law isn't just viewed as

violating the law isn't just viewed as the cost of doing business and the firm

the cost of doing business and the firm does not gain a competitive advantage

does not gain a competitive advantage through law breaking you know using ill

through law breaking you know using ill gotten gain or ill gotten data as a very

gotten gain or ill gotten data as a very basic and high level example you know

basic and high level example you know many AI models uh rely on vast amounts

many AI models uh rely on vast amounts of consumer surveillance data you know a

of consumer surveillance data you know a firm may need data just to enter the

firm may need data just to enter the market in the first instance or to best

market in the first instance or to best its competitors so firms are faced with

its competitors so firms are faced with a powerful Financial incentive to get

a powerful Financial incentive to get and use more and more consumer data in

and use more and more consumer data in order to train and enhance their models

order to train and enhance their models but if by some chance this data is

but if by some chance this data is obtained by violating any number of laws

obtained by violating any number of laws then designing a remedy that not only

then designing a remedy that not only changes the firm incentives to not do

changes the firm incentives to not do this in the future but also ensuring

this in the future but also ensuring that IL illegally obtained data does not

that IL illegally obtained data does not provide it with a Competitive Edge over

provide it with a Competitive Edge over other Market participants or stifle new

other Market participants or stifle new entry requires careful remedy design so

entry requires careful remedy design so of course you know money remedies such

of course you know money remedies such as consumer remediation and penalty are

as consumer remediation and penalty are really really important in this regard

really really important in this regard but in some instances the total remedy

but in some instances the total remedy package may need to also include more

package may need to also include more specific structural changes you know

specific structural changes you know continuing with the above an appropriate

continuing with the above an appropriate remedy May naturally include that the

remedy May naturally include that the company should be forced to delete the

company should be forced to delete the data as well but the company having the

data as well but the company having the data in the first instance may have

data in the first instance may have allowed it to already develop a new

allowed it to already develop a new model new products or have other

model new products or have other benefits so just violating the law and

benefits so just violating the law and obtaining the uh data in the first

obtaining the uh data in the first instance may have already provided with

instance may have already provided with a lot of benefits so just deleting the

a lot of benefits so just deleting the delay the data alone may not be enough

delay the data alone may not be enough the remedial package may also need to

the remedial package may also need to include um further remedies such as

include um further remedies such as algorithmic discouragement or other

algorithmic discouragement or other structural changes and you know this is

structural changes and you know this is not an exhaustive list by any means and

not an exhaustive list by any means and it's not meant to be there are many

it's not meant to be there are many different remedies that could be

different remedies that could be appropriate especially in the world of

appropriate especially in the world of AI but the point that I want to make

AI but the point that I want to make clear is that considering in designing

clear is that considering in designing remedies ensuring that you're thinking

remedies ensuring that you're thinking about the firm's business incentives the

about the firm's business incentives the root cause of the viol ation to law and

root cause of the viol ation to law and all the ways in which a firm may have

all the ways in which a firm may have benefited from the rule breaking is

benefited from the rule breaking is really important to helping to ensure

really important to helping to ensure fair and competitive markets now so far

fair and competitive markets now so far I think everything that I've said is

I think everything that I've said is basically there's no AI exception to

basically there's no AI exception to Federal Consumer Financial laws

Federal Consumer Financial laws therefore everything we're doing is the

therefore everything we're doing is the same same the one thing that I do think

same same the one thing that I do think has changed a little bit with AI and

has changed a little bit with AI and other emerging Technologies is capacity

other emerging Technologies is capacity building and one thing that's really

building and one thing that's really relevant to our AI enforcement and

relevant to our AI enforcement and remedy design work is capacity capacity

remedy design work is capacity capacity building and this is a place where the

building and this is a place where the cfpb has been really really active um

cfpb has been really really active um and specifically in

and specifically in 2022 uh we started a technologist

2022 uh we started a technologist program and what this specifically means

program and what this specifically means is that we began a program to tightly

is that we began a program to tightly embed and integrate folks with technical

embed and integrate folks with technical expertise within our supervision and

expertise within our supervision and enforcement teams amongst other areas of

enforcement teams amongst other areas of the cfpb and these are data scientists

the cfpb and these are data scientists AIML e uh experts design experts and

AIML e uh experts design experts and other technical staff and what this has

other technical staff and what this has allowed us to do is build these

allowed us to do is build these interdisciplinary teams so that we're

interdisciplinary teams so that we're approaching problems not just from the

approaching problems not just from the perspective of lawyers or economists but

perspective of lawyers or economists but rounding it out with technologist and

rounding it out with technologist and other professionals as well and this has

other professionals as well and this has really helped uh enhance our ability to

really helped uh enhance our ability to identify potential violations of law in

identify potential violations of law in the first instance gather the right

the first instance gather the right information efficiently and design

information efficiently and design meaningful remedies um and let me just

meaningful remedies um and let me just give you one example of this you know we

give you one example of this you know we all know that it's generally when you're

all know that it's generally when you're resolving a matter it's the lawyers that

resolving a matter it's the lawyers that are going to be the ones that are

are going to be the ones that are negotiating and litigating over what the

negotiating and litigating over what the appropriate remedy may be and what the

appropriate remedy may be and what the terms of that consent order might look

terms of that consent order might look like and the terms the lwes may come up

like and the terms the lwes may come up with may be beautifully written they may

with may be beautifully written they may be really precise and they make they may

be really precise and they make they may make complete sense to a non-technical

make complete sense to a non-technical audience but at the end of the day

audience but at the end of the day especially in emerging Tech especially

especially in emerging Tech especially with AI and a lot of more complex

with AI and a lot of more complex systems the person who's implementing

systems the person who's implementing that um those terms may be a non-lawyer

that um those terms may be a non-lawyer it's probably going to be a technologist

it's probably going to be a technologist a data scientist an engineer and they

a data scientist an engineer and they may have a very different understanding

may have a very different understanding of what those terms mean when they're

of what those terms mean when they're actually trying to put it into place so

actually trying to put it into place so having that extra perspective in the

having that extra perspective in the room when drafting the actual terms of

room when drafting the actual terms of the remedy so that it can be implemented

the remedy so that it can be implemented as intended it's really helpful to have

as intended it's really helpful to have these additional perspectives and that's

these additional perspectives and that's something that the cfbb has put a a big

something that the cfbb has put a a big emphasis on and I'll end my answer with

emphasis on and I'll end my answer with a a short plug that you we just closed a

a a short plug that you we just closed a posting for technologist hiring but we

posting for technologist hiring but we anticipate doing more and if anyone is

anticipate doing more and if anyone is interested go to

interested go to consumerfinance.gov and you can sign up

consumerfinance.gov and you can sign up to be uh informed about job

to be uh informed about job opportunities great so so so bright

opportunities great so so so bright lines black boxes uh thoughts from the

lines black boxes uh thoughts from the the rest of the panel on what we just

the rest of the panel on what we just heard so I I'll ask a specific question

heard so I I'll ask a specific question what um what sorts of AI use cases are

what um what sorts of AI use cases are you seeing or or do you foresee in the

you seeing or or do you foresee in the Consumer Finance sector where there's a

Consumer Finance sector where there's a potential for interaction with some of

potential for interaction with some of the bright lines already in place from

the bright lines already in place from some of these black boxes um sure I mean

some of these black boxes um sure I mean I think one is the one that I just

I think one is the one that I just discussed in credit underwriting we are

discussed in credit underwriting we are seeing you know AI being used

seeing you know AI being used alternative data being used various use

alternative data being used various use cases but I'll talk about one that's a

cases but I'll talk about one that's a little bit different we we put out an

little bit different we we put out an issue Spotlight about this um last year

issue Spotlight about this um last year and that's the growing use of chat Bots

and that's the growing use of chat Bots in banking and I think earlier today we

in banking and I think earlier today we heard um two folks talk about chatbots

heard um two folks talk about chatbots and some of them say well if you don't

and some of them say well if you don't like them too bad because they're what's

like them too bad because they're what's taking over and I think one of the

taking over and I think one of the things that this chatbot issue Spotlight

things that this chatbot issue Spotlight did is that yeah a lot of these chat

did is that yeah a lot of these chat Bots are being powered by AI some of

Bots are being powered by AI some of them are things that are claiming to be

them are things that are claiming to be AI but are much less sophisticated but

AI but are much less sophisticated but at the end of the day you know chat Bots

at the end of the day you know chat Bots are out there they're being used in

are out there they're being used in banking and one of the things that we

banking and one of the things that we did there is we actually went and looked

did there is we actually went and looked at what are consumers saying we looked

at what are consumers saying we looked at consumer complaints and you can just

at consumer complaints and you can just see complaint after complaint after

see complaint after complaint after complaint of consumers saying help I

complaint of consumers saying help I need a human I can't get help I'm stuck

need a human I can't get help I'm stuck in a doom Loop all these types of things

in a doom Loop all these types of things and we literally just search for the

and we literally just search for the word human and it's just complaint after

word human and it's just complaint after complaint after complaint you searched

complaint after complaint you searched for Doom Loop we didn't but we

for Doom Loop we didn't but we definitely uh we definitely used Doom

definitely uh we definitely used Doom Loop in the issue Spotlight but that's

Loop in the issue Spotlight but that's that is what's happening and one of the

that is what's happening and one of the things that's also really interesting is

things that's also really interesting is that you know these things they're not

that you know these things they're not just you know we talk about Ai and and

just you know we talk about Ai and and chat Bots you know hallucinating and

chat Bots you know hallucinating and it's kind of interesting I'm like to a

it's kind of interesting I'm like to a consumer it didn't hallucinate it just

consumer it didn't hallucinate it just got something wrong and the reality is

got something wrong and the reality is that there are there is a robust set of

that there are there is a robust set of rights that consumers have under a

rights that consumers have under a variety of laws consumers have um the

variety of laws consumers have um the right to have dispute resolution to have

right to have dispute resolution to have systems that can recognize that a

systems that can recognize that a consumer's trying to file a complaint or

consumer's trying to file a complaint or launch a complaint and resolve it there

launch a complaint and resolve it there are obligations to provide complete and

are obligations to provide complete and accurate information to consumers so all

accurate information to consumers so all these this entire legal framework exists

these this entire legal framework exists so sure it's not about whether or not

so sure it's not about whether or not using a chatbot or a human it's just if

using a chatbot or a human it's just if you're going to use a chat bot make sure

you're going to use a chat bot make sure you're following the law sure sure um so

you're following the law sure sure um so like to Pivot just a little bit to that

like to Pivot just a little bit to that second part of the qu what I think of

second part of the qu what I think of the second part of what I teed up

the second part of what I teed up upfront which is how to uh see this

upfront which is how to uh see this ecosystem evolve in the most Pro

ecosystem evolve in the most Pro competitive way we for examp for example

competitive way we for examp for example heard a lot about open source earlier

heard a lot about open source earlier today over the course of the day sort of

today over the course of the day sort of enabling competition but then that

enabling competition but then that raises questions of how all these open

raises questions of how all these open source firms can have the same safety

source firms can have the same safety compliance sort of protocols as really

compliance sort of protocols as really large firms and I think Shanker I think

large firms and I think Shanker I think your story and the story of trust lab is

your story and the story of trust lab is really interesting here so could you

really interesting here so could you talk a little bit about uh how firms

talk a little bit about uh how firms like yours might fit into this ecosystem

like yours might fit into this ecosystem yeah um so I think uh talking about

yeah um so I think uh talking about self-regulation right are are uh having

self-regulation right are are uh having non-regulatory ways in which uh

non-regulatory ways in which uh companies can comply um voluntarily

companies can comply um voluntarily comply uh one thing we have seen in the

comply uh one thing we have seen in the trust and safety industry is that the

trust and safety industry is that the bigger companies have actually helped

bigger companies have actually helped the smaller companies uh in terms of

the smaller companies uh in terms of providing tools and data sets uh so a

providing tools and data sets uh so a good example is uh kind of actually the

good example is uh kind of actually the worst of the internet which is child

worst of the internet which is child sexual abuse material uh so this is an

sexual abuse material uh so this is an area where uh bigger companies have

area where uh bigger companies have provided tools such as sophisticated

provided tools such as sophisticated hashing algorithms that can detect if uh

hashing algorithms that can detect if uh a certain image is present without

a certain image is present without anybody having to look at it um and also

anybody having to look at it um and also they have provided databases of hashes

they have provided databases of hashes uh because as you might imagine a bigger

uh because as you might imagine a bigger platform that has dealt with this

platform that has dealt with this problem for a long lot longer as well as

problem for a long lot longer as well as has a much bigger user base probably has

has a much bigger user base probably has dealt with like a lot more of these

dealt with like a lot more of these images uh than an upand cominging

images uh than an upand cominging platform so if they make these hashes

platform so if they make these hashes available then a small platform doesn't

available then a small platform doesn't have to go through the same problems uh

have to go through the same problems uh and can immediately uh find those issues

and can immediately uh find those issues right so and to their credit the bigger

right so and to their credit the bigger company have played their part in that

company have played their part in that another example is a coalition called

another example is a coalition called GIF City uh which was founded to counter

GIF City uh which was founded to counter uh online terrorism um and again this is

uh online terrorism um and again this is an example where bigger companies have

an example where bigger companies have provided uh tools and techniques as well

provided uh tools and techniques as well as databases that help the smaller

as databases that help the smaller companies uh which don't nearly have the

companies uh which don't nearly have the same kinds of resources uh to deal with

same kinds of resources uh to deal with uh terrorist propaganda spreading online

uh terrorist propaganda spreading online uh so those are a couple cases where

uh so those are a couple cases where where I've seen

where I've seen um with the right incentives or with the

um with the right incentives or with the right

right structure uh without even having like

structure uh without even having like necessarily

necessarily regulation uh smaller companies can

regulation uh smaller companies can benefit from the work that bigger

benefit from the work that bigger companies are already doing in the space

companies are already doing in the space uh to r with some of the risks uh now it

uh to r with some of the risks uh now it hasn't worked always though like for

hasn't worked always though like for example going past those two areas I

example going past those two areas I mean once you get into things like heat

mean once you get into things like heat speech or harassment uh things get a

speech or harassment uh things get a little more

little more graya um and and companies tend to have

graya um and and companies tend to have like somewhat different policies in that

like somewhat different policies in that regard like how much free speech uh site

regard like how much free speech uh site they lean on versus like uh more on kind

they lean on versus like uh more on kind of the user safety side um and in those

of the user safety side um and in those cases uh this hasn't worked but I wonder

cases uh this hasn't worked but I wonder if if uh with AI safety at least when we

if if uh with AI safety at least when we deal with some of the biggest risks

deal with some of the biggest risks whether similar incentives can be

whether similar incentives can be created uh where um uh kind of bigger

created uh where um uh kind of bigger companies can provide some of these

companies can provide some of these tools and techniques available uh to

tools and techniques available uh to other companies um one other thing that

other companies um one other thing that uh from our experience trust lab so we

uh from our experience trust lab so we we spoke to some of the smaller

we spoke to some of the smaller companies about uh data sharing or data

companies about uh data sharing or data pooling uh because they don't nearly

pooling uh because they don't nearly have the same like let's say a small

have the same like let's say a small platform wants to build a h hate speech

platform wants to build a h hate speech uh detection algorithm they don't nearly

uh detection algorithm they don't nearly have the same type of data that a large

have the same type of data that a large platform might have uh but across small

platform might have uh but across small platforms they might be able to pull

platforms they might be able to pull their data and and build a good system

their data and and build a good system for this um and generally speaking from

for this um and generally speaking from our experience they tend to be

our experience they tend to be interested in this but the problem is

interested in this but the problem is how do you bootstrap this uh because uh

how do you bootstrap this uh because uh obviously for for these companies the

obviously for for these companies the funding is there if there is a ready

funding is there if there is a ready benefit right whereas if it it's a sort

benefit right whereas if it it's a sort of thing where you have to collaborate

of thing where you have to collaborate on this and it may take like multiple

on this and it may take like multiple years to build the system and then you

years to build the system and then you start seeing the benefit uh it's hard to

start seeing the benefit uh it's hard to get them to invest right and not to

get them to invest right and not to mention just the coordinating such an

mention just the coordinating such an effort across many companies not easy

effort across many companies not easy right um but I wonder if the government

right um but I wonder if the government can play a role there right uh where

can play a role there right uh where they can help bootstrap similar things

they can help bootstrap similar things uh where uh smaller companies uh don't

uh where uh smaller companies uh don't have to be quite uh um uh lacking the

have to be quite uh um uh lacking the protections that a bigger company might

protections that a bigger company might put in place from safety perspective and

put in place from safety perspective and is there anything oh go ahead please

is there anything oh go ahead please well I just um I mean it's interesting

well I just um I mean it's interesting the the um Cam and terce Terror content

the the um Cam and terce Terror content hashing um I mean those are very

hashing um I mean those are very specific um kinds of content right for

specific um kinds of content right for which there is liability um and and

which there is liability um and and there was also that kind of

there was also that kind of collaboration happened in the shadow of

collaboration happened in the shadow of of government um uh action and so I

of government um uh action and so I think it's a very specific model that

think it's a very specific model that really doesn't apply to some of our

really doesn't apply to some of our other um content concerns and I think

other um content concerns and I think this also goes to you know our tour like

this also goes to you know our tour like the the law enforcement apparatus where

the the law enforcement apparatus where we have clear consumer protection laws

we have clear consumer protection laws where you can have bright line rules and

where you can have bright line rules and you have an enforcement um you know it's

you have an enforcement um you know it's a little same same with some with some

a little same same with some with some um extra capacity and that's kind of a

um extra capacity and that's kind of a built on top and that's a beautiful

built on top and that's a beautiful model but I think one of the problems is

model but I think one of the problems is that for a lot of the harms that we're

that for a lot of the harms that we're talking about with AI um you know we

talking about with AI um you know we don't yet have those that legal regime

don't yet have those that legal regime in place and to some extent we can't

in place and to some extent we can't have it in place because we have First

have it in place because we have First Amendment concerns and so you know I

Amendment concerns and so you know I think that's um

think that's um and so that militates against the legal

and so that militates against the legal certainty that um small you know

certainty that um small you know definitely small businesses want and may

definitely small businesses want and may without that legal certainty and with

without that legal certainty and with the threat of liability um they may not

the threat of liability um they may not be you know willing or able to adopt

be you know willing or able to adopt Upstream um applications so I just

Upstream um applications so I just wanted to add kind of that complexity

wanted to add kind of that complexity and so when we think about you know

and so when we think about you know regulatory regimes um or work that the

regulatory regimes um or work that the government can do in order to promote

government can do in order to promote adoption of the technology and to

adoption of the technology and to promote entry to some extent it's it's

promote entry to some extent it's it's kind of playing in that space where

kind of playing in that space where there is um uncertainty and trying to

there is um uncertainty and trying to provide you know more information more

provide you know more information more certainty um and and kind of um less

certainty um and and kind of um less risk of uh liability

overhang so let me take it down maybe one more layer um so we talked about

one more layer um so we talked about standards and developers and deployers

standards and developers and deployers there's users as well and so you know

there's users as well and so you know when we think about existing regulatory

when we think about existing regulatory regimes for technologies that are

regimes for technologies that are wonderful but dangerous uh like cars we

wonderful but dangerous uh like cars we also seek user licenses right and I

also seek user licenses right and I think the conversation earlier today

think the conversation earlier today about open source really raised the

about open source really raised the question of well could you have an open

question of well could you have an open source regime where there's a license

source regime where there's a license for the most powerful models for the

for the most powerful models for the application layer or something like that

application layer or something like that um any thoughts from the panel on

um any thoughts from the panel on whether it's user licensing or

whether it's user licensing or certification requirements at the

certification requirements at the application layer uh on regulatory

application layer uh on regulatory solution

solution down at that

level yeah I mean so I mean I'm not the expert here obviously uh don't come from

expert here obviously uh don't come from a legal background uh but one thought I

a legal background uh but one thought I had is is when I was looking at for

had is is when I was looking at for example the settlement that um I believe

example the settlement that um I believe FTC reached with uh uh with red I

FTC reached with uh uh with red I believe uh for for their use of uh AI

believe uh for for their use of uh AI technology um it it was a case where

technology um it it was a case where when when I looked at it it was a case

when when I looked at it it was a case where there was a rule already on the

where there was a rule already on the books uh which is if a compan is is

books uh which is if a compan is is employing this type of Technology uh

employing this type of Technology uh they need to have enough measures in

they need to have enough measures in place to mitigate the harms that might

place to mitigate the harms that might that might that it might cause um now

that might that it might cause um now something like that uh like the company

something like that uh like the company the various companies that are

the various companies that are developing these Solutions um they may

developing these Solutions um they may not be aware of these lws that already

not be aware of these lws that already exist on the books right um and also

exist on the books right um and also typically for the small startup I mean

typically for the small startup I mean they don't necessarily have the same

they don't necessarily have the same type of legal help uh that a bigger more

type of legal help uh that a bigger more established company might have uh

established company might have uh certainly in my B at Google when we were

certainly in my B at Google when we were dealing with gdpr compliance legal

dealing with gdpr compliance legal advice was available pretty much for

advice was available pretty much for free right so I could anytime reach out

free right so I could anytime reach out to to to the legal experts and ask them

to to to the legal experts and ask them questions um but the same is not true

questions um but the same is not true smaller companies uh so I wonder taking

smaller companies uh so I wonder taking a leap out of the the car licensing

a leap out of the the car licensing example I wonder if there if if if there

example I wonder if there if if if there could be such a thing called like a

could be such a thing called like a learner permit where maybe like there is

learner permit where maybe like there is easy to access uh materials that that

easy to access uh materials that that tell you about the rules that already

tell you about the rules that already exist on the books about like the types

exist on the books about like the types of precautions you are expected to take

of precautions you are expected to take When You're Building Technology for

When You're Building Technology for certain for certain applications and

certain for certain applications and there's many rules already on the books

there's many rules already on the books like for example for for Consumer Credit

like for example for for Consumer Credit is something that auor talked about um

is something that auor talked about um and this example that I gave uh about

and this example that I gave uh about rate so I think there easy easy to

rate so I think there easy easy to access material that um companies that

access material that um companies that are starting out in the space are

are starting out in the space are required to kind of understand before

required to kind of understand before and maybe they get a Learners permit now

and maybe they get a Learners permit now you may go ahead and start building your

you may go ahead and start building your AI models uh I mean

AI models uh I mean I'm sure like there's a lot more

I'm sure like there's a lot more thinking that needs to go to

thinking that needs to go to operationalize something like this but I

operationalize something like this but I wonder if something like that could

wonder if something like that could could could be

could could be a I have a vision of a an llm with a

a I have a vision of a an llm with a student

student driver cone on the top um other thoughts

driver cone on the top um other thoughts on this before I move on okay well I was

on this before I move on okay well I was just gonna say this this gets to some of

just gonna say this this gets to some of what elham was talking about is that um

what elham was talking about is that um the reason we can license drivers is

the reason we can license drivers is because we know what the stand we set

because we know what the stand we set the standards right we know what to um

the standards right we know what to um test them on I don't think we we would

test them on I don't think we we would know we know yet um what that would be

know we know yet um what that would be for an AI user where we have seen the

for an AI user where we have seen the kind of Licensing approach is for the

kind of Licensing approach is for the developers right for super powerful um

developers right for super powerful um flop based um standards and I think that

flop based um standards and I think that um uh Shankar you can you can uh opine

um uh Shankar you can you can uh opine on the on the um adequacy or

on the on the um adequacy or appropriateness of flop as the as the um

appropriateness of flop as the as the um measure of of capability but I've heard

measure of of capability but I've heard um you know sort of it's the best we can

um you know sort of it's the best we can do but it's not really um a very good

do but it's not really um a very good measure of um sort of risk or capability

measure of um sort of risk or capability or sort of um uh threshold so um so

or sort of um uh threshold so um so that's another place where if we were

that's another place where if we were and of course then there are the

and of course then there are the competition concerns that it locks in um

competition concerns that it locks in um uh sort of it's a gate it's a licensing

uh sort of it's a gate it's a licensing is a gatekeeping mechanism um to

is a gatekeeping mechanism um to disadvantage new entrance so that would

disadvantage new entrance so that would be um both both the measurement aspect

be um both both the measurement aspect and then also the gatekeeping aspect

and then also the gatekeeping aspect would be concerns and and what about uh

would be concerns and and what about uh enabling regulation as opposed to

enabling regulation as opposed to restrictive regulation a few people

restrictive regulation a few people today have mentioned phone number

today have mentioned phone number portability which is like one of the

portability which is like one of the great simple regulations that like just

great simple regulations that like just blew open competition in Telecom I mean

blew open competition in Telecom I mean who knew knows what would have happened

who knew knows what would have happened if you couldn't move your phone number

if you couldn't move your phone number um what about interoperability or other

um what about interoperability or other kinds of uh competition enabling sort of

kinds of uh competition enabling sort of user and a developer enabling regulatory

user and a developer enabling regulatory tools thoughts yeah um I think so as a

tools thoughts yeah um I think so as a practitioner I think like one thing that

practitioner I think like one thing that uh we have dealt with in the past couple

uh we have dealt with in the past couple years uh is

years uh is being obviously making use of great llm

being obviously making use of great llm uh technologies that have become

uh technologies that have become available but also we've also found from

available but also we've also found from time to time that we have sometimes a

time to time that we have sometimes a single point of failure dependency uh on

single point of failure dependency uh on on these uh tools and um like we had one

on these uh tools and um like we had one incident where um one of the large Lang

incident where um one of the large Lang language model providers um I think

language model providers um I think suddenly changed Its Behavior Uh and we

suddenly changed Its Behavior Uh and we our suspicion was that it was because

our suspicion was that it was because given the nature of our business because

given the nature of our business because we are trying to uh catch harmful

we are trying to uh catch harmful content the types of content that we

content the types of content that we send it uh probably their systems fig

send it uh probably their systems fig thought that like we were a bad actor

thought that like we were a bad actor trying to uh circumvent their

trying to uh circumvent their protections and so they Behavior

protections and so they Behavior immediately changed and and that meant

immediately changed and and that meant that like now we could no longer

that like now we could no longer continue offering our services the same

continue offering our services the same way right at Le we found a workaround uh

way right at Le we found a workaround uh and that was a scary moment right so and

and that was a scary moment right so and then we were like okay maybe we should

then we were like okay maybe we should investigate other models maybe they are

investigate other models maybe they are maybe they are just as powerful maybe

maybe they are just as powerful maybe they are less powerful but like we can't

they are less powerful but like we can't have a single point of failure

have a single point of failure dependency on on one provider uh but

dependency on on one provider uh but again the difficulty we had is that uh

again the difficulty we had is that uh this is a domain where the systems are

this is a domain where the systems are not very well specified right so going

not very well specified right so going back to Professor uh I guess Percy um

back to Professor uh I guess Percy um liang's comments

liang's comments in in the morning session

in in the morning session today uh the capabilities of these

today uh the capabilities of these models are not very well documented not

models are not very well documented not very well benchmarked and evaluated

very well benchmarked and evaluated right so if if you want to switch from

right so if if you want to switch from Model A to model B um without going

Model A to model B um without going through a lot of extensive testing of

through a lot of extensive testing of our own it's not easy for us to assess

our own it's not easy for us to assess um whether we can do that or not uh and

um whether we can do that or not uh and um but the type of benchmarking that he

um but the type of benchmarking that he was talking about in the morning session

was talking about in the morning session would be super help helpful right so if

would be super help helpful right so if there were independent standards that

there were independent standards that measured like different capability

measured like different capability levels of these general purpose

levels of these general purpose Technologies uh but ways in which they

Technologies uh but ways in which they more directly apply to different use

more directly apply to different use cases um because I'm aware for example

cases um because I'm aware for example there is a something like an ELO rating

there is a something like an ELO rating for measuring large language models but

for measuring large language models but that's kind of an overall capability

that's kind of an overall capability rating that doesn't tell you necessarily

rating that doesn't tell you necessarily for your specific use case how is this

for your specific use case how is this model going to compared to that other

model going to compared to that other model right um but if there were more

model right um but if there were more fine grain uh capabilities that are

fine grain uh capabilities that are being mentioned Mark like what uh the

being mentioned Mark like what uh the professor talked about in the morning uh

professor talked about in the morning uh that could really help uh from the

that could really help uh from the standpoint of not having a single point

standpoint of not having a single point of failure dependency on a single model

of failure dependency on a single model and being able to work with different uh

and being able to work with different uh providers um so that makes sense and of

providers um so that makes sense and of course interop interoperability can

course interop interoperability can apply many different layers in in in

apply many different layers in in in another session today we learned about

another session today we learned about interoperability of the hardware layer

interoperability of the hardware layer uh too so which is also pretty

uh too so which is also pretty important I was also thinking there's

important I was also thinking there's interoperability at the employee level

interoperability at the employee level so non-com

so non-com um not having non-competes is is uh an

um not having non-competes is is uh an interoperability policy yes different

interoperability policy yes different panel uh oh go ahead oh yeah I was just

panel uh oh go ahead oh yeah I was just going to say you know this really isn't

going to say you know this really isn't AI specific but more a general point I I

AI specific but more a general point I I do think that you know identifying ways

do think that you know identifying ways to help lower switching costs making it

to help lower switching costs making it easier for consumers to be able to vote

easier for consumers to be able to vote with their feet can really help Foster

with their feet can really help Foster competition Market entry you know

competition Market entry you know identifying problematic barriers to

identifying problematic barriers to entry and looking for ways to lower them

entry and looking for ways to lower them and this is one area where the cfpb is

and this is one area where the cfpb is active um you know in our personal

active um you know in our personal financial data rights proposed rul

financial data rights proposed rul making for example you know I won't get

making for example you know I won't get into the specifics of it but one of the

into the specifics of it but one of the central things that the rule is uh uh

central things that the rule is uh uh seeking to do is putting people in the

seeking to do is putting people in the driver's seat with their data so that

driver's seat with their data so that they can get better so they can get

they can get better so they can get access to better

access to better products um so in effect you know the

products um so in effect you know the proposed rule would allow folks to break

proposed rule would allow folks to break up with their Banks and that provide bad

up with their Banks and that provide bad service I.E you know help Foster Market

service I.E you know help Foster Market entry and this is by putting data back

entry and this is by putting data back into consumers uh hands and giving them

into consumers uh hands and giving them rights to ask for their data ask for

rights to ask for their data ask for their data to be shared with other

their data to be shared with other institutions sure sure I I see we just

institutions sure sure I I see we just have a few minutes left so I wanted to

have a few minutes left so I wanted to see if the audience has any questions

see if the audience has any questions before we go to the last uh lightning

before we go to the last uh lightning round where I put them on the

spot I'd love an audience question because if I don't get one I promise

because if I don't get one I promise them I'd ask a shrimp Jesus

question now they want to know what the Jesus question is so actually I'll just

Jesus question is so actually I'll just go to my lightning question in 60

go to my lightning question in 60 seconds or less to round out our panel

seconds or less to round out our panel uh what makes you hopeful about the

uh what makes you hopeful about the future of competition and regulation in

future of competition and regulation in the AI

the AI ecosystem and we'll go down the line

ecosystem and we'll go down the line just like when we started so put you on

just like when we started so put you on the spot first uh yeah I think I will go

the spot first uh yeah I think I will go with something more about from the

with something more about from the regulatory perspective which is I'm

regulatory perspective which is I'm really enthused to see and it's perhaps

really enthused to see and it's perhaps a little bit self- congratulatory but

a little bit self- congratulatory but the efforts by the US government to

the efforts by the US government to bring in more technical staff more Tech

bring in more technical staff more Tech folks with different backgrounds into

folks with different backgrounds into the room into helping us you know

the room into helping us you know prosecute cases identify remedies you

prosecute cases identify remedies you know I see Alex Gainer here from the FTC

know I see Alex Gainer here from the FTC who's also a technologist so I think

who's also a technologist so I think these programs are good and I see them

these programs are good and I see them spreading across the government really

spreading across the government really leveling up our

leveling up our work Joel caner Lena Khan and also the

work Joel caner Lena Khan and also the um uh the hu the learning that we've

um uh the hu the learning that we've seen in Congress over the past year the

seen in Congress over the past year the huge leap in understanding in Tech is

huge leap in understanding in Tech is makes me hopeful wonderful

makes me hopeful wonderful sh yeah so from my perspective a couple

sh yeah so from my perspective a couple things so one is uh I I like that there

things so one is uh I I like that there isn't just one hyperscaler there's three

isn't just one hyperscaler there's three so so at least that means some some

so so at least that means some some level of competition uh the second thing

level of competition uh the second thing I would say is also that the the pace of

I would say is also that the the pace of innovation and um the fact that

innovation and um the fact that Innovative idea can come from anywhere

Innovative idea can come from anywhere it could come from a university it could

it could come from a university it could come from like uh people like myself who

come from like uh people like myself who who quit a big tech company and then

who quit a big tech company and then started their own startup I just believe

started their own startup I just believe in the creativity uh that that we have

in the creativity uh that that we have in in in the US

in in in the US especially Alan we'll finish with you

especially Alan we'll finish with you thank you a lot of good things has been

thank you a lot of good things has been said but um I what makes me hopeful is

said but um I what makes me hopeful is is our next Generations and our younger

is our next Generations and our younger uh

uh researchers um that I see that they are

researchers um that I see that they are accepting instilling um

accepting instilling um um cultivating this culture of

um cultivating this culture of responsibility responsibility in

responsibility responsibility in technology and and and um

technology and and and um um I I told you that when I was doing

um I I told you that when I was doing machine learning I was just do it

machine learning I was just do it because I could uh now our younger

because I could uh now our younger generation uh and our researcher right

generation uh and our researcher right now uh when you look at the U uh

now uh when you look at the U uh criteria for Acceptance in papers in NS

criteria for Acceptance in papers in NS and others thinking about impact and

and others thinking about impact and impact is one of those so getting our

impact is one of those so getting our younger um next generation of you know

younger um next generation of you know researchers thinking about the impact of

researchers thinking about the impact of the technology as part of the design of

the technology as part of the design of the research as part of the experiment

the research as part of the experiment and um their better awareness of shared

and um their better awareness of shared accountability and responsibility uh is

accountability and responsibility uh is something that keeps me helpful

something that keeps me helpful wonderful and and I am hopeful because

wonderful and and I am hopeful because experts like you are so willing to

experts like you are so willing to engage and talk as has happened all day

engage and talk as has happened all day so uh hold your applause for them for a

so uh hold your applause for them for a moment uh I just want to note that AAG

moment uh I just want to note that AAG caner will be back up to share a few

caner will be back up to share a few concluding remarks uh before we get to

concluding remarks uh before we get to the reception but thanks again to this

the reception but thanks again to this wonderful panel

don't you go first and then I'll okay great so um thanks everyone

great so um thanks everyone congratulations on making it to the

congratulations on making it to the halfway point um so we're really excited

halfway point um so we're really excited for the remainder of the program it's

for the remainder of the program it's going to be riveting um this was really

going to be riveting um this was really a warm-up um wait until till you see

a warm-up um wait until till you see salmon Jesus um uh and then clam sorry

salmon Jesus um uh and then clam sorry um so I don't want to blab her this has

um so I don't want to blab her this has been an extraordinary conversation and I

been an extraordinary conversation and I I couldn't do it justice with with a

I couldn't do it justice with with a wrap up other than to say I very much

wrap up other than to say I very much appreciate Dave um finishing on the note

appreciate Dave um finishing on the note of what makes people hopeful and um what

of what makes people hopeful and um what makes me hopeful is um conversations

makes me hopeful is um conversations like um the discussion today um

like um the discussion today um demonstrates that when we bring together

demonstrates that when we bring together people from different points of view um

people from different points of view um uh we can start to actually move forward

uh we can start to actually move forward in solving um or identifying important

in solving um or identifying important questions and then which is a necessary

questions and then which is a necessary first step to solving it and I think um

first step to solving it and I think um one of the features of today and I'm

one of the features of today and I'm really happy with the way it worked was

really happy with the way it worked was that um we we came together here uh in

that um we we came together here uh in in pal Alto um and and at Stanford and

in pal Alto um and and at Stanford and we heard from people who have technology

we heard from people who have technology background investment background um we

background investment background um we heard from people who focus on Hardware

heard from people who focus on Hardware we focus on data machine learning um but

we focus on data machine learning um but we also heard uh from content creators

we also heard uh from content creators and people who have that perspective and

and people who have that perspective and I think um one of the ways we organized

I think um one of the ways we organized today's discussion and I'm glad that we

today's discussion and I'm glad that we did is uh thinking about the stack from

did is uh thinking about the stack from top to bottom um and I think that's an

top to bottom um and I think that's an important uh framing which is what are

important uh framing which is what are the different components of the stack

the different components of the stack how do we make sure that there's

how do we make sure that there's competition within each of those

competition within each of those components and then competition is

components and then competition is enabled and how those various components

enabled and how those various components interoperate uh and that really is from

interoperate uh and that really is from everything from the the chip down to the

everything from the the chip down to the to the N user I I think it would be

to the N user I I think it would be important though it's very important um

important though it's very important um to

to remember that uh humans are part of that

remember that uh humans are part of that stack uh and whether it's the labor of

stack uh and whether it's the labor of Engineers uh or the C uh the creations

Engineers uh or the C uh the creations of of content creators um artists

of of content creators um artists journalists um actors

journalists um actors um

um musicians uh poets um or just plain old

musicians uh poets um or just plain old people um that's an not only an

people um that's an not only an important input into um the creation of

important input into um the creation of of AI but it is an essential input um

of AI but it is an essential input um and making sure that we uh breathe life

and making sure that we uh breathe life into um a healthy competitive ecosystem

into um a healthy competitive ecosystem that returns um results for all in a way

that returns um results for all in a way that uh allows everybody to benefit is

that uh allows everybody to benefit is so important and so I think that's what

so important and so I think that's what makes me hopeful as we can come away

makes me hopeful as we can come away from today's conversation and think

from today's conversation and think about how do we how do we work together

about how do we how do we work together to enable that so without further Ado

to enable that so without further Ado I'll turn it over and um thank everybody

I'll turn it over and um thank everybody for their participation and attendance

for their participation and attendance thank

you so just to close out on behalf of Stanford I'd like to thank you for being

Stanford I'd like to thank you for being in the room and thank for everyone who's

in the room and thank for everyone who's online and still stuck with us today um

online and still stuck with us today um we will have today's recordings

we will have today's recordings available on our our website in addition

available on our our website in addition we've been doing one-on-one interviews

we've been doing one-on-one interviews with a number of panelists and those

with a number of panelists and those will also be featured on our website in

will also be featured on our website in a few maybe give us a little bit few

a few maybe give us a little bit few weeks to pull those together um but stay

weeks to pull those together um but stay tuned for that and um I'd also just like

tuned for that and um I'd also just like to thank our events team and our

to thank our events team and our videographing video video team and

videographing video video team and support just for all their great work

support just for all their great work today thank you so much we'll um meet

today thank you so much we'll um meet you all back outside at the reception

you all back outside at the reception and thank you again

Click on any text or timestamp to jump to that moment in the video

Most transcripts ready in under 5 seconds

One-Click Copy125+ LanguagesSearch ContentJump to Timestamps

Paste YouTube URL

Enter any YouTube video link to get the full transcript

Most transcripts ready in under 5 seconds

Get Our Chrome Extension

Get transcripts instantly without leaving YouTube. Install our Chrome extension for one-click access to any video's transcript directly on the watch page.

Add to Chrome — Free

Works with YouTube, Coursera, Udemy and more educational platforms

Get Instant Transcripts: Just Edit the Domain in Your Address Bar!

YouTube

←

→

↻

https://www.youtube.com/watch?v=UF8uR6Z6KLc

YoutubeToText

←

→

↻

https://youtubetotext.net/watch?v=UF8uR6Z6KLc

YouTube Transcript:Promoting Competition in AI

Video Transcript

Paste YouTube URL

Transcript Extraction Form

Get Our Chrome Extension

Get Instant Transcripts: Just Edit the Domain in Your Address Bar!

YouTube Transcript:
Promoting Competition in AI