0:03 I found that for many job roles people
0:07 have a mental picture of that iconic TSS
0:10 that uniquely defines that job R for
0:12 example computer programmers right code
0:15 doctors maybe see patients lawyers go to
0:17 court to argue court cases and I think
0:18 that when people think about AI
0:21 opportunities often is instinctual to
0:25 say Can AI do that most iconic role or
0:28 that most iconic task of that job but I
0:29 found that when we actually systematic
0:32 an ize the task that the particular job
0:34 is made up of the best opportunities may
0:36 or may not be that first initial
0:38 Instinct let's take a look at a few
0:40 examples if you look at the task the
0:42 computer programmer does they do write
0:44 code and so it's tempting when thinking
0:46 about AI for computer programming to ask
0:49 can AI help write code but it turns out
0:51 computer programmers do many other
0:53 things they have the right documentation
0:55 they sometimes respond to user support
0:58 requests they often review others codes
0:59 and they often gather request
1:01 requirements for what a piece of
1:04 software is intended to do and if you
1:06 were to evaluate the generative AI
1:10 potential for this job you may find that
1:12 writing codes can be help of AI but it's
1:15 a relatively difficult task but maybe
1:18 writing documentation is actually easier
1:21 to do with Gen of AI and so on don't
1:23 take the Gena of AI potential column to
1:25 seriously in these examples since these
1:28 are informal evaluations and if were to
1:30 do a rigorous evaluation based on
1:33 technical feasibility and business value
1:35 your specific conclusions may be
1:38 different but I think that is actually
1:41 easier to get gen of AI to write
1:43 documentation for code than to actually
1:45 write the code itself and in many
1:47 different job rowes the best potential
1:50 for AI may not be the most obvious first
1:53 TS you might think of let's look at
1:55 another example lawyers spend a lot of
1:58 time Drafting and reviewing legal
2:00 documents they often have to to answer
2:02 client's questions on how to interpret
2:05 laws if preparing for a court case
2:07 they'll have to review evidence and
2:08 sometimes they're involved in
2:10 negotiating settlements and sometimes to
2:13 represent clients in court and I find
2:15 that a systematic listing out of these
2:17 toss as well as the systematic
2:19 evaluation of the potential May
2:22 sometimes lead to interesting
2:25 conclusions so I think that there's a
2:27 high potential of regen of AI to help
2:29 with Drafting and reviewing legal
2:31 documents as well as maybe with
2:34 interpreting laws whereas I can't see a
2:36 lawyer sending a robot to court to argue
2:38 on their behalf at least not for some
2:40 time and so if you work with the law
2:42 firm and Analysis like this might help
2:45 you decide where you actually want to
2:48 use genter VII one more example
2:51 Landscaping a landscaper has to maintain
2:53 and care for plants purchase and
2:55 transport plants maintain equipment Comm
2:57 with clients maintain a Business website
2:59 and so on I'm listing of course just a
3:01 subset of the TSS that any of these job
3:04 rules do if you were to do an analysis
3:06 yourself you may end up with anywhere
3:10 from five to 15 to 30 toss per job row
3:13 and in this case I think most of these
3:16 toss actually have pretty low genive AI
3:18 potential and so the work of a
3:21 landscaper may be less impacted in the
3:23 next few years by genitive AI compared
3:27 to computer programmers and lawyers so
3:29 that's how you can analyze jobs by
3:32 breaking it down into tasks and I
3:33 encourage you to think through what are
3:36 the tasks in your work and where gvi may
3:38 be able to help or for the business you
3:40 may be involved in to think about how
3:42 gender VII could help many different
3:44 tasks in that
3:46 business when people think about
3:49 augmentation or automation people's
3:51 minds often go initially to cost savings
3:53 because if you automate something seems
3:56 like you can you know save money but in
3:59 most ways of Technology Innovation going
4:00 back all the way way to say the
4:02 invention of the steam engine to
4:04 electricity to the computer many
4:06 companies started off thinking about
4:08 cost savings but ended up actually
4:10 putting even more of the effort into
4:13 pursuing Revenue growth and that's
4:16 because growth has no limit but you know
4:19 you can only save so much money and when
4:22 certain tasks are automated it turns out
4:24 sometimes you can rethink the workflow
4:26 of how the business creates value so for
4:28 example if you could do something a
4:30 thousand times cheaper because of
4:34 automation say answering queries from
4:36 customers then rather than just taking
4:38 the cost savings you may be able to
4:41 build a new type of customer service
4:44 organization that serves people a
4:46 thousand times better and this type of
4:47 thinking can lead to growth
4:50 opportunities that go well beyond cost
4:51 savings let's take a look at some
