0:03 last week I spent 5 hours and $49 to
0:05 complete Google's latest AI Essentials
0:07 course for beginners and since I need to
0:08 recoup that money to fund my addiction
0:11 to parasocial relationships with AI I
0:13 made this video to share five key
0:15 takeaways from the course the pros and
0:18 cons and to give you a definitive answer
0:19 as to whether this certificate you
0:22 receive at the end will indeed get you
0:25 laid I mean paid paid more because you
0:27 now have a new skill that was a weird
0:30 slip of tongue there takeaway number one
0:32 broadly speaking there are three types
0:34 of AI tools out there first Standalone
0:36 tools are AI powered software designed
0:39 to work independently with minimal setup
0:41 this category includes general purpose
0:43 chat Bots like chache BT Gemini Claude
0:46 and perplexity as well as specialized
0:49 apps like spico otter AI mid journey and
0:51 Gamma although they serve completely
0:53 different needs all these tools are
0:55 classified as Standalone because they
0:57 can be accessed directly through their
1:00 websites or apps and can be used without
1:03 integration with other software this is
1:05 in contrast to the second category tools
1:08 with integrated AI features which refers
1:10 to built-in enhancements within a
1:13 particular piece of software for example
1:15 after drafting a post in Google Docs I
1:17 can either copy and paste this text over
1:20 to a standalone app like chbt to improve
1:23 my writing or I can use the built-in or
1:26 integrated Gemini for workspace AI
1:28 feature to make adjustments another
1:30 example would be I can either use mid
1:32 Journey as a standalone tool to generate
1:34 images for a presentation or I can
1:37 create an image directly within Google
1:39 Slides by using again gemni for
1:42 workspace in these instances chbt and M
1:44 Journey are Standalone AI tools and
1:47 Google Docs and Google slides are tools
1:49 with integrated AI features third a
1:52 custom AI solution is an application
1:54 that's tailor made to solve a specific
1:56 problem for instance John Hopkins
1:59 University developed an AI system with
2:02 the sole objective of detecting sepsis
2:04 this custom AI solution improved
2:06 diagnostic accuracy from 2 to 5% to an
2:09 average of 40% and if you're like me and
2:10 have no technical background you might
2:12 think oh custom AI Solutions are
2:14 extremely Technical and I'll probably
2:15 never have to use them in the workplace
2:18 in reality the opposite is true because
2:20 well-designed custom AI Solutions should
2:23 require little to no technical knoow for
2:25 example when I was in the sales team I
2:27 managed over 200 clients every quarter
2:29 and Performing research on every single
2:32 one was obviously timec consuming
2:33 nowadays there are custom AI solutions
2:35 that can ingest all the information
2:37 about those 200 clients taking into
2:39 account factors like seasonality
2:42 historical data and Industry Trends and
2:44 rank those clients by How likely they
2:46 are to need assistance helping the
2:49 salesperson prioritize their time by the
2:51 way if you're thinking of buying
2:53 Google's AI Essentials course don't
2:54 because I only found out after I paid
2:56 for it you get the AI Essentials course
2:59 for free if you enroll in the Google
3:01 project management certification on
3:03 corsera who is kindly sponsoring this
3:05 portion of the video regular viewers
3:06 know that I have a full-time job and
3:08 project management obviously plays a
3:10 large part in my day-to-day to be
3:12 transparent though I'm mainly
3:14 self-taught since there wasn't a go-to
3:16 course back in the day but I recently
3:17 started the project management
3:20 certification on corsera since it's now
3:22 like the golden standard for project
3:24 management project management is
3:26 literally applicable to all Industries
3:28 and roles so if you want to be more
3:29 organized in the workplace click the
3:31 link below to enroll in the Google
3:33 project management certification and
3:36 unlock the AI Essentials course for free
3:38 thank you corsera for sponsoring this
3:40 part of the video back to my takeaways
3:41 the second learning from the course is a
3:43 prompt engineering tip and it's to
3:45 surface the implied context to
3:48 illustrate this imagine your vegetarian
3:49 friend asks you for restaurant
3:52 recommendations you instinctively reply
3:55 with vegetarian friendly options even if
3:57 your friend doesn't explicitly say hey
3:59 idiot make sure it's vegetarian here
4:01 here the fact that your friend is a
4:04 vegetarian is implied context and needs
4:06 to be explicitly stated when
4:08 communicating with AI tools like Chachi
4:11 BT and Google Gemini another example
4:13 might be you're preparing to negotiate a
4:15 raise with your boss you know in your
4:18 head that last year you received a 10%
4:20 increase this year you're the highest
4:22 performer on the team and the industry
4:25 average is a 12% increase so you decide
4:28 to ask for 15% raise If You Leave Out
4:30 All That implied context when
4:32 brainstorming negotiation techniques
4:35 with an AI tool you will receive a lower
4:38 quality AKA more generic output if you
4:40 want to dive deeper into this I have an
4:41 entire video on how to write the perfect
4:43 prompt so I'll leave a link to that down
4:45 below and if you want to make a copy of
4:47 my five favorite prompts for
4:49 productivity I'll leave a link to my
4:51 completely free workspace toolkit as
4:53 well takeaway number three know when to
4:55 use zero shot and few shot prompting in
4:58 a nutshell the word shot simply means
5:00 examples zero shot means you use a
5:03 prompt with no examples one shot means
5:06 you include one example few shot means
5:08 you include two or more examples for
5:11 instance a zero shot example might look
5:12 something like write me a pickup line
5:15 for Bumble which is a completely
5:16 hypothetical scenario I would never
5:19 condone much less participate in a
5:21 one-hot prompt would be write me a
5:22 pickup line for Bumble reference this
5:24 pickup line my friend used that worked
5:26 well for him and you include an example
5:28 of what your friend wrote A F shop
5:30 prompt would look the same as one shot
5:32 with just two or more examples of
5:34 successful pickup lines and as you can
5:36 imagine the more relevant examples you
5:38 provide the AI tool the more relevant
5:41 the output and for the record if my H
5:43 future wife is watching this I've
5:45 obviously never done this myself this is
5:47 just an example for this video as a
5:49 matter of fact I don't I don't even use
5:52 dating apps takeaway number four use
5:54 Chain of Thought prompting for complex
5:56 tasks I've talked about this concept in
5:58 previous videos but I really like this
5:59 simple and straightforward definition
6:02 from Google's course when you divide a
6:05 single task into more manageable steps
6:07 you help the large language model
6:09 produce accurate and consistent results
6:11 a relatable real word example would be
6:13 writing a cover letter option one you
6:15 share your current resume and the job
6:16 description with the chatbot and just
6:19 prompt it to write you a cover letter
6:20 option two with Chain of Thought
6:22 prompting you break the large task
6:24 writing an entire cover letter into
6:26 manageable steps with step one being
6:27 based on my resume and the job
6:29 description write an attention Gra
6:32 having hook for my cover letter then
6:34 after making some minor tweaks to the
6:36 hook paragraph step two would be pasting
6:38 that back into the chatbot and asking it
6:41 for the body paragraph rinse and repeat
6:43 for the closing paragraph if you want
6:45 proof that this works I have an entire
6:46 video where I show job Seekers how to
6:48 use Chain of Thought prompting to not
6:50 only write cover letters but to also
6:52 improve their resumés I'll leave a link
6:54 down below takeaway number five is a
6:56 topic that many of us including myself
6:59 tend to overlook when using AI tools and
7:02 it's to understand the limitations of AI
7:03 broadly speaking there are three main
7:06 limitations first the underlying data
7:08 that is used to train AI models may be
7:11 biased if a text to image model is only
7:13 trained on minimalistic Graphics it
7:15 might not be able to produce flashy and
7:18 bold designs second there simply wasn't
7:20 enough information in the source
7:22 training data on a given topic many AI
7:24 models have a cutoff date so if you ask
7:26 for something that happened recently it
7:28 won't have enough data on that topic to
7:30 give you an accurate accurate answer
7:32 finally hallucinations are AI outputs
7:34 that are straight up factually
7:36 inaccurate sometimes this is a feature
7:38 and not a bug when you're brainstorming
7:40 new ideas for example other times
7:42 hallucinations perpetuate false
7:44 information so you definitely want to
7:46 double check your answer for what I call
7:48 high stake tasks for example what sort
7:50 of supplement or vitamin to take given
7:52 your health goals when it comes to pros
7:53 and cons I'm going to start with who
7:56 this course is not for this course is
7:58 not for you if you're already using AI
8:00 tools like chpt and Google Gemini as
8:02 part of your daily workflow and you're
8:05 looking to dive deeper into specific AI
8:07 use cases and that's because although
8:09 they do a great job explaining complex
8:11 topics the examples they give in the
8:14 course are pretty vague for instance in
8:16 one of the lessons they gave an example
8:19 where a company uses AI to decrease
8:22 customer service response times and that
8:24 was it that was entire example it would
8:26 have been nice to dive a bit deeper and
8:28 talk about whether this company used a
8:31 standalone AI tool or custom AI solution
8:33 how they trained the workers to use the
8:37 AI system how they grounded the data so
8:39 that the AI system didn't hallucinate
8:41 since this was a customer service um
8:43 application right so stuff like that
8:45 that being said this is an excellent
8:47 beginner level course with three huge
8:49 advantages first you're learning from
8:51 Google employees who are established
8:54 experts in the field of AI they know
8:55 what they're talking about and they're
8:56 not just some random person on the
9:01 internet who enjoys making crass jokes
9:03 second as a visual learner I'm amazed at
9:05 how they're able to use Simple Graphics
9:08 to explain complex topics for example
9:11 they analogize AI tools and AI models to
9:14 a car and its engine the model sleng
9:16 engine provides the underlying
9:19 capabilities while the tool like the car
9:22 helps you in completing the task third
9:23 the interactive elements are
9:25 surprisingly helpful the activities or
9:28 homework are well-designed in that they
9:29 actually help you learn a key concept
9:31 from that lesson the graded assignments
9:33 are also not that easy they're mainly
9:35 multiple choice but you actually have to
9:37 pay attention to get 80% and pass the
9:39 quiz last but not least the course
9:42 provides a curated list of AI tools for
9:44 beginners to explore and includes a
9:47 glossery of common AI terms that are now
9:50 prevalent in our daily lives to sum up
9:51 this course is great for beginners
9:53 visual Learners and can use the legit
9:55 certificate you get the end to attract
9:58 prospective employers and or Partners um
10:00 if you this video helpful you might want
10:02 to check out my summary of Google's free
10:05 AI course that's a bit more conceptual
10:07 but equally important in the meantime
10:08 though as
