davidshapiro_youtube_transcripts
/
Generative AI for CEOs How to think about AI today and how to get the most ROI from it_transcript.csv
text,start,duration | |
good morning everybody David Shapiro,0.12,4.56 | |
here with another video so today's video,2.34,4.2 | |
is going to be a little bit less about,4.68,4.98 | |
the narratives around Ai and that sort,6.54,4.139 | |
of stuff and it's going to be a little,9.66,4.019 | |
bit more uh down to Basics uh let's talk,10.679,5.641 | |
let's talk business with artificial,13.679,5.041 | |
intelligence uh particularly generative,16.32,3.48 | |
AI,18.72,3.3 | |
uh real quick plug before we get started,19.8,5.04 | |
uh I am available for one-on-one,22.02,6.179 | |
consultations on patreon and I'm also,24.84,6.3 | |
available uh for for different kinds of,28.199,4.981 | |
engagements uh reach out to me on,31.14,4.259 | |
LinkedIn if you'd like to talk about any,33.18,3.18 | |
of the things that I'm talking about,35.399,2.34 | |
here,36.36,3.48 | |
all right so getting right into the show,37.739,5.361 | |
let's talk about generative AI for CEOs,39.84,5.28 | |
and we're going to look at it at a very,43.1,3.88 | |
high level so first we're going to look,45.12,3.18 | |
at it through the lens of people,46.98,4.919 | |
processes and tools uh all organizations,48.3,5.759 | |
go through Transformations and one of,51.899,4.14 | |
the ways to look at organizational,54.059,4.621 | |
transitions is through people processes,56.039,5.281 | |
and tools what kinds of people do you,58.68,4.44 | |
need and how does that how does that,61.32,4.14 | |
change what kind of expertise training,63.12,4.56 | |
that sort of stuff how do your processes,65.46,4.38 | |
change uh within the business and then,67.68,5.04 | |
what tools do you use so this is just,69.84,4.92 | |
kind of the the the background framing,72.72,4.02 | |
we're not going to unpack these in any,74.76,3.719 | |
particular order we're just going to,76.74,3.96 | |
kind of talk about generative AI through,78.479,3.96 | |
this particular lens,80.7,3.54 | |
so one thing that you can think about,82.439,4.741 | |
generative AI is that it's just more,84.24,3.96 | |
software,87.18,3.6 | |
at a fundamental level generative AI is,88.2,4.559 | |
just a new kind of software but it is,90.78,3.9 | |
still just software,92.759,4.921 | |
one thing to keep in mind is that uh,94.68,5.52 | |
technology you know computers uh,97.68,4.14 | |
software that sort of stuff has,100.2,3.48 | |
automated and accelerated business for,101.82,4.02 | |
decades this is nothing new it's nothing,103.68,4.399 | |
different this is just the next flavor,105.84,5.4 | |
for those of you in the technology space,108.079,4.921 | |
you might remember how virtualization,111.24,4.44 | |
was a very disruptive technology for the,113.0,5.32 | |
data center and that ultimately led to,115.68,4.52 | |
uh the the,118.32,4.74 | |
preeminence of cloud computing which is,120.2,4.959 | |
now pretty much standard and so the,123.06,5.22 | |
combination of virtualization and apis,125.159,6.001 | |
and stuff that made cloud computing uh,128.28,4.98 | |
you know the viable business model that,131.16,4.68 | |
it is today this is no different from,133.26,5.119 | |
how Oracle and SQL Technologies,135.84,4.74 | |
transformed uh the way that businesses,138.379,4.481 | |
handled data and then of course there's,140.58,4.32 | |
everything that Microsoft did for for,142.86,3.66 | |
business productivity from their office,144.9,4.02 | |
suite to Microsoft Windows and,146.52,4.26 | |
everything else team Foundation,148.92,3.959 | |
everything,150.78,3.36 | |
so,152.879,3.061 | |
from that perspective from a software,154.14,4.98 | |
perspective you can think of a,155.94,6.06 | |
generative AI in two primary categories,159.12,5.399 | |
the first category is that generative AI,162.0,4.62 | |
is a new kind of automation,164.519,4.621 | |
and so you know again business,166.62,6.3 | |
automation nothing new uh and business,169.14,5.879 | |
automation is also it doesn't really,172.92,4.74 | |
change the idea is that you just do more,175.019,4.681 | |
with less you do it faster and you do it,177.66,5.52 | |
more reliably so looking at generative,179.7,6.179 | |
AI through the lens of automation the,183.18,4.98 | |
idea is that it can take some of the,185.879,4.621 | |
workload off of humans and put it onto,188.16,4.98 | |
machines and then it can allow you to,190.5,5.879 | |
get more done faster with less and more,193.14,5.28 | |
reliably that is the entire point of,196.379,3.72 | |
automation,198.42,3.66 | |
another thing that you can look at it is,200.099,3.661 | |
and this is kind of the fundamentally,202.08,5.1 | |
new aspect is it has to do with decision,203.76,5.88 | |
enhancement and what I mean by that is,207.18,4.76 | |
that the particularly language models,209.64,5.459 | |
they can Aid with all kinds of things,211.94,6.04 | |
whether it's judgment creativity,215.099,6.0 | |
appraisal brainstorming drafting they,217.98,5.399 | |
can also read a tremendous amount in,221.099,3.601 | |
very short periods of time which can,223.379,3.78 | |
allow you to make better decisions by,224.7,5.58 | |
ingesting more data than ever before,227.159,5.22 | |
so by thinking of generative Ai and,230.28,5.459 | |
these two uh like I guess categories,232.379,5.94 | |
Automation and decision enhancement that,235.739,4.741 | |
can really help you target where in your,238.319,3.78 | |
corporation where in your organization,240.48,5.52 | |
to look at using generative AI again the,242.099,5.521 | |
idea of automation is that you get more,246.0,4.04 | |
done faster and of course time is money,247.62,4.56 | |
another thing is that you can do it with,250.04,3.46 | |
less you might be able to do it with,252.18,3.479 | |
less head count or you can take the,253.5,3.72 | |
ideally you take the head count that you,255.659,2.58 | |
do have because we don't want to see,257.22,3.239 | |
anyone get laid off if we can avoid it,258.239,5.46 | |
and then you you Empower your existing,260.459,6.3 | |
staff to do twice as much four times as,263.699,6.0 | |
much 10 times as much so that is and,266.759,4.141 | |
then of course you can also use,269.699,3.961 | |
generative AI to perform checks and,270.9,5.22 | |
interlocks which can raise the quality,273.66,5.4 | |
of work and make it more consistent so,276.12,4.799 | |
those are a few uh kind of off the cuff,279.06,3.12 | |
ways to think about it like yes,280.919,3.901 | |
generative AI is just more software but,282.18,4.68 | |
there's it does give us a couple new,284.82,4.26 | |
capabilities of this software,286.86,4.98 | |
So speaking of new capabilities uh,289.08,5.28 | |
generative AI mostly right now is text,291.84,4.38 | |
and images but there's other modalities,294.36,5.22 | |
coming soon namely video audio and a few,296.22,5.64 | |
other things so what are the what are,299.58,4.38 | |
the Baseline capabilities that this,301.86,4.5 | |
technology has first of all as a,303.96,4.92 | |
language model it is primarily based on,306.36,5.16 | |
text which means it can read everything,308.88,4.94 | |
it can ingest email chat papers,311.52,4.86 | |
scientific papers articles laws,313.82,5.26 | |
protocols reports logs tribal knowledge,316.38,5.759 | |
in the form of KB articles documentation,319.08,4.5 | |
that sort of stuff,322.139,3.241 | |
so when you think when you have a,323.58,4.08 | |
machine that can read at a thousand,325.38,5.22 | |
times faster than any human think of the,327.66,4.8 | |
text Heavy aspects of your business,330.6,4.5 | |
whether that is combing through,332.46,5.76 | |
financial reports or consolidating,335.1,6.18 | |
Knowledge from from chat logs customer,338.22,5.46 | |
feedback there's all kinds of stuff that,341.28,3.96 | |
is text based or could easily be,343.68,3.78 | |
rendered as text that you can then just,345.24,4.5 | |
read and get a tremendous amount of,347.46,5.16 | |
information from now the problem there,349.74,4.32 | |
of course is you have to be asking the,352.62,3.72 | |
right questions and there are also still,354.06,4.62 | |
technical limits namely the window size,356.34,3.72 | |
of some of these things,358.68,4.019 | |
you can summarize and transform text and,360.06,4.02 | |
information and so what I mean by that,362.699,3.121 | |
is you can take a large block of text,364.08,4.86 | |
and you can either make it smaller or,365.82,4.68 | |
change the formatting or that sort of,368.94,4.02 | |
thing which again this is one of the two,370.5,4.8 | |
one of the new tools in the toolbox,372.96,4.92 | |
that allows you to engage with the data,375.3,4.679 | |
that you do have in a different manner,377.88,4.62 | |
uh brainstorming and planning so this is,379.979,5.16 | |
one of the most powerful aspects of,382.5,5.16 | |
tools such as chat gbt it's like hey I'm,385.139,4.201 | |
addressing this problem help me,387.66,4.2 | |
brainstorm how to approach it uh I've,389.34,4.979 | |
used brainstorming everything for uh,391.86,4.32 | |
writing fiction you know it's like hey,394.319,4.201 | |
I'm working on this scene uh help me,396.18,4.079 | |
brainstorm a few ways to approach the,398.52,3.959 | |
scene so that it you know ticks these,400.259,4.321 | |
boxes right you give it you give it some,402.479,4.5 | |
constraints and Technologies like chat,404.58,4.679 | |
GPT will absolutely surprise you in,406.979,4.141 | |
terms of what they're capable of helping,409.259,5.041 | |
you brainstorm because they it has a lot,411.12,4.859 | |
more information than than you will ever,414.3,3.3 | |
have and so if you give it the right,415.979,3.181 | |
constraints and the right problem space,417.6,3.78 | |
it will surprise you and it'll add,419.16,4.56 | |
information to uh what you're trying to,421.38,4.5 | |
do planning one of the things that's,423.72,3.96 | |
really good at is helping you think,425.88,4.08 | |
through every step of a plan,427.68,4.859 | |
brainstorming and planning are among the,429.96,4.679 | |
most cognitively demanding tasks and,432.539,4.44 | |
we'll talk about cognitive demands and,434.639,4.921 | |
cognitive offload in just a moment but,436.979,4.62 | |
basically you know there are some things,439.56,3.9 | |
that chat GPT and similar Technologies,441.599,4.621 | |
can do that can save you a tremendous,443.46,4.92 | |
amount of mental energy which then you,446.22,4.379 | |
can allocate to other things other,448.38,4.5 | |
problems that AI is not yet capable of,450.599,4.021 | |
doing and then finally Drafting and,452.88,4.379 | |
revising uh you know whether you're,454.62,4.44 | |
writing a legal document or an agreement,457.259,4.621 | |
or a contract or even just drafting an,459.06,5.28 | |
email I mean I use chat GPD it's like,461.88,4.74 | |
hey I need to respond to this email I,464.34,3.9 | |
want to do it diplomatically and explain,466.62,4.32 | |
it clearly and you know like yes I am,468.24,4.799 | |
good at writing but it's just easier to,470.94,4.199 | |
have the machine write the first draft,473.039,5.1 | |
for me and then I can clean it up and,475.139,4.921 | |
then revisions as well,478.139,3.661 | |
in some of my previous videos you might,480.06,4.44 | |
have seen where I use uh the chat GPT,481.8,4.98 | |
API to provide feedback on my fiction,484.5,4.8 | |
and and plenty of other things I often,486.78,5.16 | |
use chat GPT for uh helping actually,489.3,5.22 | |
even make these slide decks although I'm,491.94,3.84 | |
getting better at making these slide,494.52,3.239 | |
decks so I actually need it less which,495.78,3.539 | |
is a really interesting phenomenon it's,497.759,3.66 | |
like okay I've actually gained new,499.319,4.201 | |
skills by using this tool and I am,501.419,3.601 | |
better at communicating now and I need,503.52,3.66 | |
this tool less really fascinating I,505.02,3.299 | |
don't know if that's going to keep up,507.18,4.079 | |
but anyway so here these are four basic,508.319,6.001 | |
categories of new capabilities that you,511.259,5.041 | |
can think of in terms of okay what is it,514.32,3.899 | |
that generative AI is actually capable,516.3,4.44 | |
of right now and then from there these,518.219,3.721 | |
are kind of the low level building,520.74,4.02 | |
blocks that you can slot into other,521.94,5.22 | |
products other services whether it's,524.76,4.139 | |
internal or external right because you,527.16,4.02 | |
might have internal facing tools you,528.899,4.44 | |
might have external facing tools and,531.18,3.42 | |
these are things that you can just kind,533.339,3.241 | |
of pipe into existing Frameworks,534.6,4.56 | |
existing processes to make them better,536.58,4.5 | |
you can also think about these when,539.16,4.56 | |
creating new products and services,541.08,5.66 | |
okay so as I promised cognitive offload,543.72,6.72 | |
most humans are capable of two to maybe,546.74,5.86 | |
four hours of high performance cognitive,550.44,5.16 | |
labor per day many people less so an,552.6,4.56 | |
average a rule of thumb is about two,555.6,2.76 | |
hours,557.16,4.2 | |
so the idea of cognitive offload is that,558.36,4.56 | |
it's like okay whatever it is that's,561.36,4.8 | |
difficult you uh delegate that to,562.92,5.34 | |
someone else you delegate it to a,566.16,3.66 | |
machine,568.26,3.68 | |
excuse me you delegate it to a machine,569.82,5.22 | |
you uh you do something to make it,571.94,6.399 | |
easier you create a process or you,575.04,5.4 | |
offshore it or whatever,578.339,3.661 | |
so what are some things that are,580.44,3.74 | |
cognitively demanding decision making,582.0,4.38 | |
there's a phenomenon called decision,584.18,5.44 | |
fatigue if you follow hbr uh and or,586.38,5.579 | |
scitec daily or whatever there's plenty,589.62,3.899 | |
of research that goes into decision,591.959,3.421 | |
fatigue and basically what this means is,593.519,4.921 | |
that the cognitive demand of making,595.38,5.7 | |
decisions all day is very very taxing,598.44,4.68 | |
it's very draining and so what happens,601.08,4.379 | |
is that after making too many decisions,603.12,5.1 | |
once decision fatigue takes over people,605.459,4.32 | |
will just kind of go with the default,608.22,3.299 | |
answer or whatever they're most,609.779,3.961 | |
comfortable with or they'll use simpler,611.519,4.5 | |
and simpler heuristics and objectively,613.74,4.98 | |
speaking decisions get worse with with,616.019,5.701 | |
time and fatigue so any place that you,618.72,6.66 | |
can use generative AI to offload some of,621.72,5.46 | |
the decision-making process from humans,625.38,4.019 | |
or to augment it right because it,627.18,3.719 | |
doesn't necessarily mean taking it from,629.399,4.201 | |
Human but instead instead of having to,630.899,4.981 | |
have a human synthesize all the things,633.6,4.14 | |
and think through it you can have the AI,635.88,3.959 | |
generator report like okay based on the,637.74,3.3 | |
problem space that you're addressing,639.839,3.721 | |
here's the top three choices you know,641.04,4.14 | |
think through this and again this is,643.56,3.24 | |
something that I use chat GPT for all,645.18,3.36 | |
the time it's like okay help me think,646.8,3.06 | |
through what you know what are the,648.54,2.82 | |
questions that I'm forgetting to ask,649.86,4.38 | |
myself that sort of thing and by by,651.36,4.32 | |
having the machine do some of the,654.24,3.659 | |
thinking for me it preserves my,655.68,4.2 | |
cognitive energy for other problems,657.899,4.56 | |
speaking of problem solving,659.88,5.28 | |
problem solving is also very very,662.459,5.461 | |
cognitively demanding whether you're,665.16,4.44 | |
trying to solve a financial problem a,667.92,3.2 | |
legal problem a technology problem,669.6,4.02 | |
having something that you can bounce,671.12,5.5 | |
ideas off of or ideally actually just,673.62,5.58 | |
pipe it into your existing products and,676.62,5.219 | |
services and platforms where either it,679.2,4.8 | |
can solve the problem for you or it can,681.839,4.68 | |
help you solve problems again this is,684.0,6.12 | |
going to be a high value ad where the,686.519,4.981 | |
overall intelligence of your,690.12,3.6 | |
organization will go up if you use,691.5,4.2 | |
generative AI to help your organization,693.72,4.619 | |
solve problems better but it also save,695.7,4.74 | |
your employees brain power for the,698.339,3.901 | |
problems that AI is not yet capable of,700.44,2.7 | |
solving,702.24,3.779 | |
learning learning is something that uh,703.14,4.62 | |
All Humans Do it's natural it's,706.019,3.841 | |
instinctive it's reflective sorry,707.76,3.9 | |
reflexive you don't even need to think,709.86,3.9 | |
about doing it but when you put yourself,711.66,3.9 | |
in a position where you're deliberately,713.76,3.84 | |
learning and stretching your mind that,715.56,4.56 | |
is very exhausting it's very taxing uh,717.6,4.62 | |
and so what you can do is you can use,720.12,5.459 | |
these tools to help streamline your,722.22,5.16 | |
training your learning,725.579,4.741 | |
and what that does is that you uh can,727.38,4.68 | |
meet everyone where they are for your,730.32,4.38 | |
training programs and you can also often,732.06,4.5 | |
learn faster because then you have an,734.7,3.66 | |
interactive modality and you can,736.56,3.899 | |
basically create an internal uh,738.36,4.68 | |
customized tutor that says okay here's,740.459,4.44 | |
the company's training material let's,743.04,3.539 | |
get through it and then instead of just,744.899,3.781 | |
like watching a brain dead boring video,746.579,3.961 | |
that nobody actually cares about anyways,748.68,4.62 | |
you can have an interactive experience,750.54,4.799 | |
which one will make the training more,753.3,4.8 | |
engaging uh and that that level of,755.339,4.921 | |
Engagement means that people will enjoy,758.1,4.739 | |
it more and they'll retain it better,760.26,4.44 | |
and then of course that's you know,762.839,3.3 | |
corporate training is one thing but,764.7,3.42 | |
there's also there's also the uh,766.139,4.921 | |
acquisition of new skills so let's say,768.12,4.56 | |
you're bringing you're onboarding a new,771.06,3.839 | |
product or a new service or you're,772.68,3.719 | |
shipping a new product and service and,774.899,3.481 | |
you need to teach people about it so,776.399,3.541 | |
remember that learning is very,778.38,3.899 | |
cognitively demanding and AI chat Bots,779.94,4.62 | |
are already good at tutoring and,782.279,4.441 | |
teaching and so this again is a way,784.56,4.079 | |
where you can use the AI not to replace,786.72,4.32 | |
humans but to augment humans and bring,788.639,3.781 | |
them to the next level,791.04,4.56 | |
creativity synthesizing new ideas is,792.42,5.099 | |
also very very difficult it's very,795.6,4.979 | |
cognitively demanding and while,797.519,5.461 | |
Technologies like chat GPT and others,800.579,4.26 | |
are not technically creating anything,802.98,3.78 | |
new most humans don't create anything,804.839,3.901 | |
new anyways we just remix stuff that,806.76,3.72 | |
we've experienced which is not any,808.74,3.42 | |
different from what generative AI does,810.48,4.14 | |
and also within the context of business,812.16,4.2 | |
you don't necessarily need to create,814.62,5.04 | |
anything entirely 100 new what you do,816.36,5.34 | |
need to do is take the things that you,819.66,5.46 | |
know and then deploy them with with good,821.7,6.12 | |
application you need to find the right,825.12,4.32 | |
solution for the problem that you're,827.82,2.639 | |
facing,829.44,3.36 | |
whether it's a marketing problem legal,830.459,4.201 | |
Finance HR right if you need to come up,832.8,4.86 | |
with a creative solution for an HR issue,834.66,6.119 | |
again chat GPT knows enough about the,837.66,5.28 | |
human condition and emotions and all,840.779,3.3 | |
this other stuff to help you think,842.94,2.82 | |
through and come up with Creative,844.079,4.26 | |
Solutions uh whatever it is that you're,845.76,4.68 | |
doing and then finally planning as I,848.339,4.201 | |
mentioned planning is very cognitively,850.44,3.959 | |
demanding and this is this includes,852.54,4.32 | |
forecasting anticipation or even,854.399,4.801 | |
predicting the future and so basically,856.86,3.84 | |
one thing that you can do is you can ask,859.2,5.16 | |
chat gbt say like hey uh you know the,860.7,5.22 | |
these are the business trends that I'm,864.36,3.479 | |
seeing this is what the Market's doing,865.92,3.96 | |
help me think through this and it it's,867.839,3.901 | |
not necessarily going to be 100 right,869.88,4.199 | |
but that's not the point what I'm saying,871.74,3.96 | |
is you're not using it to replace your,874.079,3.06 | |
ability to plan you're using it to,875.7,3.66 | |
augment your ability to plan,877.139,4.38 | |
so the golden rule for all of this one,879.36,4.14 | |
of the best ways you can use generative,881.519,4.26 | |
AI this Golden Rule is preserve human,883.5,4.68 | |
cognition for when you really need it so,885.779,5.641 | |
this design principle goes uh it one,888.18,5.88 | |
it's it's relatively old this is one of,891.42,4.56 | |
the chief design principles in the Unix,894.06,4.2 | |
operating system which treats human,895.98,5.34 | |
cognition as a scarce resource but also,898.26,5.28 | |
this is how jet fighters work so the,901.32,5.1 | |
avionics research industry uh for,903.54,5.58 | |
particularly for military realized that,906.42,4.919 | |
you needed to treat a pilot's cognitive,909.12,4.86 | |
attention as the scare as a very scarce,911.339,5.401 | |
resource and so the avionics in jet,913.98,5.76 | |
fighters only gets their attention when,916.74,4.86 | |
they absolutely need it,919.74,4.08 | |
and so if you watch the top gun Maverick,921.6,3.479 | |
movie,923.82,2.879 | |
um one of the scenes where the plane is,925.079,3.661 | |
like spiraling towards the ground and it,926.699,3.721 | |
the alarm starts where it's like you,928.74,4.14 | |
know pull up pull up pull you know and,930.42,4.68 | |
then as the uh as the level of danger,932.88,5.759 | |
Rises the alarm changes right and so,935.1,6.419 | |
that change in tone uh what that does is,938.639,4.5 | |
it calls the Pilot's attention to,941.519,3.721 | |
something that is at that they need to,943.139,4.56 | |
pay attention to immediately uh and so,945.24,5.159 | |
that is an example of how uh attention,947.699,5.221 | |
engineering can be used for good rather,950.399,3.781 | |
than the way that it's used in social,952.92,4.08 | |
media and this is just an example but if,954.18,5.94 | |
you the the whole point here is that the,957.0,5.579 | |
the point of cognitive offload is to,960.12,5.1 | |
treat your employees brain power those,962.579,4.38 | |
those two hours of high performance,965.22,3.66 | |
cognitive labor as a very scarce,966.959,4.62 | |
resource so that instead of you know,968.88,5.1 | |
using their raw brain power to do the,971.579,4.981 | |
work manually instead they use a machine,973.98,4.38 | |
to offload as much of that as possible,976.56,4.2 | |
so that then they can protect their,978.36,4.919 | |
productivity for or elsewhere,980.76,4.92 | |
so here's some questions to ask so these,983.279,4.201 | |
are based on questions that I often ask,985.68,5.219 | |
when I'm Consulting uh and this is these,987.48,5.099 | |
are the questions that I ask to zoom,990.899,4.62 | |
into and find okay where is it that you,992.579,5.161 | |
need to use generative Ai and how do you,995.519,4.44 | |
employ it so the first question what,997.74,3.959 | |
activities are the most menial tedious,999.959,4.201 | |
and time consuming what do you avoid,1001.699,5.161 | |
doing because it's just too painful,1004.16,5.7 | |
so this is a really good question I,1006.86,4.38 | |
usually start with something like this,1009.86,3.539 | |
especially if if uh you know my client,1011.24,4.08 | |
or whatever doesn't really know what,1013.399,3.36 | |
they need or know how to deploy,1015.32,3.0 | |
generative AI it's like okay what are,1016.759,3.481 | |
your pain points and of course this is a,1018.32,3.66 | |
business practice that is as old as,1020.24,3.719 | |
business itself what are your pain,1021.98,3.66 | |
points let's see if we can address those,1023.959,4.62 | |
pain points with generative AI,1025.64,4.98 | |
and you know the the way that it can,1028.579,3.6 | |
address it is you know look at cognitive,1030.62,3.36 | |
offload and new capabilities,1032.179,3.841 | |
the second question is what activities,1033.98,4.62 | |
actually add value what is the,1036.02,4.559 | |
objectively uh beneficial business,1038.6,3.78 | |
output of a given activity or in other,1040.579,4.141 | |
words what is the value proposition what,1042.38,3.959 | |
is what is the activity that you're,1044.72,3.66 | |
doing that is actually giving something,1046.339,4.321 | |
to your customers what are you getting,1048.38,4.14 | |
paid to do what is the outcome that,1050.66,3.84 | |
they're looking for because if you apply,1052.52,4.26 | |
generative AI directly to your outputs,1054.5,4.22 | |
to the things that bring in Revenue,1056.78,4.68 | |
obviously that's good for business what,1058.72,4.3 | |
a lot of startups are doing right now is,1061.46,3.48 | |
trying to invent new products and new,1063.02,4.38 | |
services to create entirely new revenue,1064.94,4.08 | |
streams but there's plenty of existing,1067.4,4.2 | |
businesses out there that could probably,1069.02,5.159 | |
help their Top Line you know get the get,1071.6,5.66 | |
the net uh income up by,1074.179,6.24 | |
providing more goods and services either,1077.26,5.919 | |
at a cheaper price or adding some novel,1080.419,4.341 | |
features to their products and services,1083.179,4.5 | |
but by by actually paying attention to,1084.76,6.159 | |
the objective business value of any,1087.679,4.981 | |
activity and this it doesn't necessarily,1090.919,3.541 | |
have to be the business outputs it could,1092.66,3.78 | |
be the outputs of a given Department the,1094.46,3.839 | |
HR department your legal department your,1096.44,3.42 | |
it Department your marketing department,1098.299,3.601 | |
right because one of the what is the,1099.86,3.78 | |
output of the marketing department you,1101.9,3.899 | |
generate you generate leads which can,1103.64,3.419 | |
lead to revenue,1105.799,3.12 | |
so that's what I mean by outputs what is,1107.059,4.681 | |
the objective value added number three,1108.919,5.161 | |
how do you know that an activity is,1111.74,3.42 | |
successful,1114.08,4.68 | |
whether it's a kpi intuition experience,1115.16,5.7 | |
what is it that you actually pay,1118.76,4.5 | |
attention to to know that you have been,1120.86,5.22 | |
successful the reason that this is a,1123.26,5.159 | |
good question for generative AI is,1126.08,4.44 | |
because sometimes it's not obvious how,1128.419,3.961 | |
generative AI can help you be more,1130.52,4.56 | |
successful so if they're like you know,1132.38,4.26 | |
if you're looking for an intangible,1135.08,3.42 | |
right like maybe you need to build,1136.64,4.56 | |
customer trust or build rapport that is,1138.5,5.7 | |
something that uh that might surprise,1141.2,4.56 | |
you but chat GPT and other similar,1144.2,3.0 | |
Technologies can actually help you with,1145.76,5.159 | |
you say like uh hey I you know I'm gonna,1147.2,6.78 | |
I've got a customer that I'm cultivating,1150.919,4.801 | |
um let me feed you in the profile for,1153.98,3.54 | |
this customer let's talk about this,1155.72,3.18 | |
customer and let's connect with what,1157.52,3.36 | |
they actually need right how can I how,1158.9,3.659 | |
can I make this customer trust me how,1160.88,4.08 | |
can I earn their trust right and so,1162.559,4.98 | |
that's just one example of where if it's,1164.96,5.7 | |
a marketing a sales sales team,1167.539,5.661 | |
it can generative AI can help them,1170.66,5.519 | |
increase their level of success if you,1173.2,4.599 | |
pay attention to what it is that,1176.179,3.481 | |
actually makes them successful and it,1177.799,3.12 | |
does not necessarily have to be,1179.66,2.639 | |
something that you can put a number on,1180.919,3.961 | |
in fact a lot of business is something,1182.299,4.141 | |
that is very difficult to measure,1184.88,5.1 | |
objectively now that being said chat GPT,1186.44,6.0 | |
doesn't necessarily need metrics this is,1189.98,3.9 | |
one of the biggest differences between,1192.44,2.94 | |
conventional machine learning and,1193.88,3.36 | |
artificial intelligence and generative,1195.38,4.86 | |
AI it operates in qualitative space not,1197.24,5.16 | |
quantitative space this is the biggest,1200.24,4.679 | |
fundamental difference between data,1202.4,4.019 | |
science in the past and data science,1204.919,4.441 | |
today is that pivot from quantitative,1206.419,5.76 | |
data numbers forecast linear regression,1209.36,5.04 | |
that sort of thing to qualitative,1212.179,4.201 | |
information what is it that this person,1214.4,4.019 | |
likes and why and how do you earn their,1216.38,4.5 | |
trust and finally question number four,1218.419,4.861 | |
what activities are the highest priority,1220.88,5.34 | |
or highest stakes in other words where,1223.28,4.8 | |
would having an extra brain or an extra,1226.22,4.26 | |
set of eyes make the most difference,1228.08,5.64 | |
so this is something where uh often if,1230.48,4.439 | |
something is high priority or high,1233.72,3.42 | |
stakes that is where that's like the,1234.919,4.201 | |
sweet spot right that is that is the,1237.14,3.18 | |
meat and potatoes of your business,1239.12,3.66 | |
because that is the that is the set of,1240.32,5.099 | |
tasks that only humans can do right now,1242.78,5.1 | |
and they are they're going to be the,1245.419,4.62 | |
bottleneck right so what I mean by that,1247.88,4.38 | |
is uh I used to work at managed service,1250.039,5.401 | |
providers and so you know one of the,1252.26,6.18 | |
biggest things was uptime if the if the,1255.44,5.099 | |
data center goes down or a service goes,1258.44,4.619 | |
down that's money out the door so that,1260.539,4.26 | |
means that's a very high priority task,1263.059,4.021 | |
and so if I were Consulting for a,1264.799,3.601 | |
managed service provider and they said,1267.08,3.54 | |
Dave how can we use generative AI to,1268.4,3.84 | |
help our bottom line I said what are the,1270.62,3.72 | |
what are the biggest uh priorities that,1272.24,4.2 | |
you have and it's like well if we have a,1274.34,4.44 | |
severity one let's you know we need to,1276.44,5.04 | |
all hands on deck and one thing this is,1278.78,4.139 | |
just from my experience in technology,1281.48,3.24 | |
one thing is that a lot of people are,1282.919,3.241 | |
very reactive,1284.72,3.42 | |
I actually had to leave a job because my,1286.16,4.259 | |
boss did not understand the value of,1288.14,4.5 | |
eliminating technical debt and so he's,1290.419,4.441 | |
just like oh well we have fire alarms on,1292.64,4.019 | |
a daily basis and that's just the way,1294.86,3.179 | |
things are and I said no the hell it,1296.659,3.661 | |
isn't like prevention is what we should,1298.039,4.441 | |
be doing right if if your fire,1300.32,3.66 | |
department is over taxed because,1302.48,3.48 | |
everything is burning down maybe you,1303.98,4.199 | |
need new fire code he didn't get that so,1305.96,4.98 | |
I left that job but my point is is that,1308.179,5.041 | |
there there are going to be key risks,1310.94,4.8 | |
and key business activities that are,1313.22,4.86 | |
responsible for a good chunk of the,1315.74,4.26 | |
money coming in and the money going out,1318.08,3.599 | |
so that's what I mean by high priority,1320.0,3.84 | |
and high stakes you pay attention to,1321.679,3.961 | |
those things first and you fix them as,1323.84,4.98 | |
much as you can by adding generative AI,1325.64,6.0 | |
uh to make you know to to change those,1328.82,4.44 | |
ratios so they got more coming in and,1331.64,4.68 | |
less going out uh you know you can ask,1333.26,5.88 | |
and as a CEO you can ask these questions,1336.32,5.46 | |
of every department head you can also,1339.14,4.32 | |
look at it from an internal perspective,1341.78,3.66 | |
or an external perspective how do you,1343.46,3.719 | |
engage with your customer your vendors,1345.44,3.96 | |
your clients how do you engage with your,1347.179,3.901 | |
internal stakeholders how do you engage,1349.4,3.659 | |
with your uh your financial stakeholders,1351.08,3.3 | |
your shareholders,1353.059,4.681 | |
these are you can generative AI can,1354.38,4.799 | |
apply to literally every level of,1357.74,2.88 | |
business,1359.179,4.38 | |
okay so on the tool side so we talked,1360.62,4.679 | |
about the people in the processes let's,1363.559,3.24 | |
talk about the tools,1365.299,4.86 | |
uh first one thing that you can do the,1366.799,5.041 | |
best one of the approaches is that you,1370.159,4.741 | |
can create drop-in tools so one thing is,1371.84,4.8 | |
that a lot of people get wrong is,1374.9,3.24 | |
they're trying is they try and build,1376.64,3.6 | |
tools that require a fundamentally,1378.14,3.779 | |
different workflow and it's a,1380.24,2.88 | |
fundamentally different approach to,1381.919,3.841 | |
business and I this is something that I,1383.12,4.799 | |
often talk about with uh for startups,1385.76,3.84 | |
where they're like oh we've got this,1387.919,3.301 | |
great new tool that we're building it's,1389.6,3.48 | |
you know based on generative Ai and I'm,1391.22,4.02 | |
like okay great but how are businesses,1393.08,4.92 | |
going to use it is this gonna is is this,1395.24,5.4 | |
gonna require them to adopt an entirely,1398.0,4.86 | |
new workflow is it going to replace,1400.64,5.94 | |
existing tools or ideally it's a drop-in,1402.86,5.76 | |
tool that just goes in alongside every,1406.58,4.26 | |
other tool in their toolbox so for,1408.62,3.419 | |
instance in the marketing department,1410.84,3.54 | |
they already use file servers they,1412.039,5.161 | |
already use Photoshop they use Adobe and,1414.38,5.159 | |
and image services,1417.2,4.92 | |
stock image services so in that case,1419.539,4.5 | |
image generators like stable diffusion,1422.12,3.98 | |
and mid-journey and adobe's Firefly,1424.039,5.221 | |
these are all drop-in tools where they,1426.1,4.959 | |
fit into the existing workflows,1429.26,4.44 | |
perfectly and so when you're whether,1431.059,4.62 | |
you're looking at adopting a generative,1433.7,4.859 | |
AI tool or service make sure that it,1435.679,5.701 | |
will fit into your existing workflows,1438.559,4.081 | |
because that's where you're going to get,1441.38,3.179 | |
the most value out of it if it's a tool,1442.64,3.419 | |
that requires a fundamentally different,1444.559,4.5 | |
approach the the path to adoption is,1446.059,5.641 | |
just that much harder and I you know I,1449.059,4.801 | |
saw this on the back in my it,1451.7,4.38 | |
infrastructure days all the time where,1453.86,3.42 | |
it's like yeah there's a great new,1456.08,3.42 | |
powerful tool but like we can't really,1457.28,3.84 | |
use it because it doesn't integrate with,1459.5,3.419 | |
anything else that we do and it's going,1461.12,3.36 | |
to fundamentally change the way we do,1462.919,3.601 | |
business which just means it's not a,1464.48,4.199 | |
good fit and this is this is something,1466.52,4.92 | |
for startup CEOs to keep in mind it's,1468.679,5.521 | |
also something for CEOs that are not,1471.44,6.119 | |
Tech Centric or AI Centric when adopting,1474.2,4.8 | |
AI tools,1477.559,3.181 | |
make sure that you make sure that you,1479.0,4.32 | |
get those drop-in tools one another,1480.74,3.9 | |
thing to keep in mind is that general,1483.32,3.3 | |
purpose tools like chat GPT are often,1484.64,4.26 | |
the hardest to use because you have to,1486.62,4.02 | |
know a lot about the tool you have to,1488.9,3.36 | |
know a lot about what it is that you're,1490.64,3.06 | |
trying to do and you have to interact,1492.26,2.94 | |
with it in a very fundamentally,1493.7,2.64 | |
different way than you've ever,1495.2,4.14 | |
interacted with any uh technology tool,1496.34,4.199 | |
in the past,1499.34,4.199 | |
so because of that part of what makes a,1500.539,5.041 | |
a drop-in tool is going to be the,1503.539,4.02 | |
affordances and user experience design,1505.58,4.56 | |
that says oh this is just like this,1507.559,4.62 | |
other piece of software that I use this,1510.14,4.14 | |
other app that I use all the time it,1512.179,4.5 | |
just has some AI baked into it and it's,1514.28,4.2 | |
really easy to use this is one of the,1516.679,4.201 | |
things that I consulted with in the,1518.48,3.48 | |
early days,1520.88,2.52 | |
was there was a lot of people just,1521.96,4.14 | |
putting together generic dashboards of,1523.4,4.62 | |
AI tools and I'm like you have no,1526.1,4.86 | |
coherent strategy there's nothing about,1528.02,5.399 | |
this that just stands out and says Ah I,1530.96,5.099 | |
know exactly what this tool is for and,1533.419,5.221 | |
you know so basically what you're trying,1536.059,4.441 | |
to optimize for is make your generative,1538.64,4.32 | |
AI tool as intuitive as possible where,1540.5,3.9 | |
you can just pick someone off the street,1542.96,3.42 | |
put them in front of the app and they,1544.4,5.159 | |
say oh I know I one I know exactly what,1546.38,4.74 | |
this does and I know exactly how to use,1549.559,4.5 | |
it just by guessing if you can if you,1551.12,4.86 | |
can do that then you uh almost certainly,1554.059,3.901 | |
have a drop in tool,1555.98,3.9 | |
now the other thing that you can do is,1557.96,4.56 | |
you can improve existing systems so this,1559.88,4.679 | |
could be you know if you've got teams if,1562.52,3.72 | |
you've got slack if you've got exchange,1564.559,4.561 | |
you know or or even log services or,1566.24,4.439 | |
whatever and of course that's just all,1569.12,4.2 | |
looking at it from the I.T side there's,1570.679,4.521 | |
plenty of other business platforms,1573.32,4.38 | |
whether it's your financial workflows,1575.2,4.5 | |
your marketing workflows,1577.7,5.16 | |
whether if you if you if you're not,1579.7,5.32 | |
trying to add a new tool what you can do,1582.86,5.28 | |
is if you have access to the code base,1585.02,4.98 | |
uh for you know your internally,1588.14,4.8 | |
developed tools you can add generative,1590.0,6.419 | |
AI uh for at some of these uh decision,1592.94,5.94 | |
points in order to make it just a little,1596.419,4.74 | |
bit better a little bit faster so the,1598.88,3.779 | |
primary example I have in my head is,1601.159,3.481 | |
servicenow and other,1602.659,3.781 | |
um internal management things like that,1604.64,4.5 | |
so let's say you have ticket routing,1606.44,5.16 | |
you've got uh follow-ups you've got,1609.14,4.86 | |
emails so basically like you know,1611.6,5.4 | |
instead of having uh humans do a lot of,1614.0,5.159 | |
the work you can have the machine do a,1617.0,3.539 | |
lot of the work for the humans and to,1619.159,3.421 | |
say hey which one of these emails do you,1620.539,4.26 | |
want to send to this customer as a,1622.58,5.28 | |
follow-up uh or if there's a if there's,1624.799,4.98 | |
a ticketing system you can have the,1627.86,4.26 | |
ticketing system ask a few obvious,1629.779,4.26 | |
questions before a human even looks at,1632.12,4.2 | |
it so that there's by the time your,1634.039,3.901 | |
ticket gets escalated to an actual,1636.32,4.5 | |
engineer or whoever it gets to a few,1637.94,4.2 | |
questions have already been asked,1640.82,3.42 | |
immediately and so there's more context,1642.14,5.279 | |
which can reduce your mttx,1644.24,5.34 | |
another place that you can look at it is,1647.419,5.76 | |
human intervention where in uh and it,1649.58,5.04 | |
can either be routing to the correct,1653.179,2.821 | |
human so I was actually just talking to,1654.62,3.48 | |
a friend about this on the medical side,1656.0,4.2 | |
uh who's at a startup trying to use,1658.1,4.559 | |
generative AI which is uh if there's a,1660.2,4.32 | |
if there's a patient need how do you,1662.659,3.661 | |
write it to the route it to the correct,1664.52,4.38 | |
medical provider that is a that is a,1666.32,4.2 | |
great use case because you can feed a,1668.9,4.62 | |
whole patient chart into the llm and say,1670.52,5.039 | |
what what exactly competency what,1673.52,3.84 | |
competencies does this person need right,1675.559,2.761 | |
now,1677.36,2.34 | |
um and then you get the you get the,1678.32,2.82 | |
exact right thing and then you make sure,1679.7,4.14 | |
that uh that that uh you know nurse or,1681.14,5.519 | |
or specialist also gets the exact notes,1683.84,5.219 | |
that they need uh without having to go,1686.659,4.14 | |
talk to a you know a bunch of other,1689.059,4.321 | |
people because I mean heck one of my one,1690.799,4.321 | |
of my patreons several of my patreon,1693.38,3.6 | |
supporters are in the medical field and,1695.12,3.48 | |
they talk about just how annoying it is,1696.98,3.72 | |
when charts aren't updated correctly and,1698.6,4.559 | |
when uh Physicians and nurses don't talk,1700.7,4.2 | |
to each other or they you know forget to,1703.159,4.26 | |
pass something on so again you can,1704.9,5.58 | |
reduce human interaction uh or sorry you,1707.419,4.14 | |
can reduce the need for human,1710.48,3.12 | |
intervention by having some of these,1711.559,3.901 | |
interactions automated,1713.6,4.38 | |
and then finally cognitive offload if,1715.46,4.079 | |
you if you go hunting across your,1717.98,2.939 | |
organization for the things that are,1719.539,3.781 | |
really demanding those are ways that you,1720.919,4.38 | |
can improve existing systems because you,1723.32,4.02 | |
know you reduce the frustration that,1725.299,3.661 | |
will increase productivity increase,1727.34,3.9 | |
satisfaction and that also just all,1728.96,4.26 | |
speaks to the bottom line,1731.24,4.5 | |
okay so that was people processes and,1733.22,4.74 | |
tools let's talk about better faster and,1735.74,4.919 | |
cheaper otherwise uh how do you get the,1737.96,4.319 | |
best return on investment from any,1740.659,3.361 | |
generative AI initiative whether it's a,1742.279,2.941 | |
product or service that you're thinking,1744.02,2.759 | |
about adopting or one that you're,1745.22,3.54 | |
building uh internally,1746.779,4.801 | |
so the the primary three questions here,1748.76,5.34 | |
that you should ask yourself is uh what,1751.58,4.44 | |
can generative AI do better than humans,1754.1,4.14 | |
what can it do faster than humans and,1756.02,4.32 | |
what can it do cheaper than humans and,1758.24,3.419 | |
so this is what I mean where it's like,1760.34,3.719 | |
okay if you can offset some of the labor,1761.659,5.821 | |
costs uh you can ideally hopefully you,1764.059,5.401 | |
preserve your existing head count but,1767.48,3.6 | |
you take your organization to the next,1769.46,3.48 | |
level and you can double triple,1771.08,4.62 | |
quadruple productivity while also doing,1772.94,4.859 | |
cognitive offload so your your human,1775.7,4.74 | |
employees are happier,1777.799,4.98 | |
so this is a universal truth of,1780.44,4.8 | |
technology technology has always been a,1782.779,5.4 | |
force multiplier so you know the the,1785.24,5.22 | |
Industrial Revolution example that I,1788.179,3.961 | |
always love to give is the tractor,1790.46,3.54 | |
tractors are very complicated they're,1792.14,4.019 | |
expensive heavy pieces of equipment but,1794.0,4.32 | |
we still use them because they have the,1796.159,4.62 | |
physical power output of literally,1798.32,5.459 | |
hundreds of horses uh and so you get rid,1800.779,4.681 | |
of the you instead of feeding and,1803.779,3.78 | |
providing veterinary care for hundreds,1805.46,4.14 | |
of horses you have a tractor that all it,1807.559,4.201 | |
does is drink gasoline need some oil and,1809.6,3.54 | |
some replacement parts every now and,1811.76,2.22 | |
then,1813.14,3.96 | |
generative AI is a cognitive Force,1813.98,5.52 | |
multiplier so let me say that again the,1817.1,4.26 | |
the first industrial or the first three,1819.5,4.399 | |
industrial revolutions were more about,1821.36,5.28 | |
uh communication Force multiplication,1823.899,5.461 | |
and physical Force multiplication,1826.64,6.24 | |
generative AI is the first example of a,1829.36,5.679 | |
cognitive Force multiplier it doesn't,1832.88,4.14 | |
just send email faster right because you,1835.039,3.721 | |
know email came around in the 80s or,1837.02,3.48 | |
whatever which is like okay cool,1838.76,3.899 | |
communication now happens faster that's,1840.5,5.34 | |
not really cognitive labor or crunching,1842.659,5.281 | |
numbers right math definitely very,1845.84,4.38 | |
cognitively demanding but this is a new,1847.94,4.56 | |
era of of cognitive labor force,1850.22,5.699 | |
multiplication so that's why this video,1852.5,5.88 | |
has cognitive offload as a central,1855.919,4.86 | |
lesson as something to look for when,1858.38,4.26 | |
you're trying to integrate generative AI,1860.779,3.841 | |
into your business,1862.64,2.94 | |
better,1864.62,4.88 | |
so uh keep your ear to the ground but,1865.58,7.02 | |
you know basically chat GPT has already,1869.5,5.5 | |
demonstrated Superior uh clinical,1872.6,4.38 | |
judgment than many doctors and this is,1875.0,4.5 | |
according to Harvard research it also,1876.98,4.559 | |
has better uh executive judgment,1879.5,4.98 | |
executive reasoning uh than many people,1881.539,5.301 | |
and this includes managers and some CEOs,1884.48,5.22 | |
and so in this case look for the,1886.84,4.66 | |
activities where generative AI is,1889.7,5.28 | |
demonstrably superior to humans this,1891.5,4.919 | |
will give you obviously a competitive,1894.98,3.12 | |
Advantage but if it is better than you,1896.419,4.86 | |
use the machine right humans can go use,1898.1,5.28 | |
a rake and a hoe and and and and and,1901.279,4.921 | |
plow uh the field but we don't have,1903.38,4.5 | |
humans doing it by hand anymore because,1906.2,3.599 | |
the machine is better so just because,1907.88,3.6 | |
you can do something doesn't mean that,1909.799,3.781 | |
you should especially if the machine is,1911.48,4.02 | |
better than you at it now again this,1913.58,4.56 | |
doesn't necessarily mean get handed over,1915.5,4.98 | |
to the machine entirely but you should,1918.14,4.38 | |
be using the machine to augment your,1920.48,5.1 | |
abilities so in the case of doctors if,1922.52,6.18 | |
there is any room for doubt maybe,1925.58,5.219 | |
doctors should be required or maybe not,1928.7,5.099 | |
required but you know as part of the the,1930.799,4.801 | |
medical resourcing system like you know,1933.799,3.541 | |
epic and Cerner and stuff,1935.6,3.72 | |
you integrate some uh some language,1937.34,3.66 | |
model technology into that where it's,1939.32,3.78 | |
like okay cool uh based on this,1941.0,4.08 | |
patient's chart and test results have,1943.1,4.38 | |
you thought of X Y and Z uh just enough,1945.08,4.14 | |
just enough to clue in that the,1947.48,3.54 | |
physician on the direction that maybe,1949.22,4.199 | |
they need to go in order to augment,1951.02,4.379 | |
human judgment and augment human,1953.419,4.62 | |
executive reasoning now one thing about,1955.399,6.481 | |
this is that unless you have a you know,1958.039,5.701 | |
a product or service where they've done,1961.88,4.5 | |
the work and they can say look this this,1963.74,4.62 | |
AI service that we've built is,1966.38,4.26 | |
demonstrably better than humans you're,1968.36,3.48 | |
going to need to figure that out for,1970.64,2.639 | |
yourself through experimentation and,1971.84,3.3 | |
consultation,1973.279,2.581 | |
um,1975.14,3.0 | |
and also as I mentioned earlier you will,1975.86,3.6 | |
need to start measuring some things,1978.14,3.539 | |
qualitatively rather than quantitatively,1979.46,5.76 | |
so if if the objective is patient,1981.679,6.061 | |
satisfaction or customer satisfaction or,1985.22,4.559 | |
trust or any number of those other,1987.74,4.08 | |
intangible things they can be a lot,1989.779,4.14 | |
harder to measure but fortunately,1991.82,4.5 | |
language models can help you measure,1993.919,5.461 | |
those things through approximations,1996.32,4.8 | |
and so what I mean by that is just ask,1999.38,3.24 | |
chat GPT,2001.12,2.939 | |
um you know how can I measure customer,2002.62,3.779 | |
satisfaction or trust or whatever and,2004.059,3.901 | |
it'll give you some cool ideas about how,2006.399,3.481 | |
you can measure that and then you say,2007.96,2.939 | |
well what's the difference between,2009.88,3.179 | |
quantitative and qualitative trust me if,2010.899,3.481 | |
you do that experiment you'll see what,2013.059,2.401 | |
I'm talking about,2014.38,3.539 | |
uh but again this is no different from,2015.46,4.559 | |
using machines to do things that humans,2017.919,5.1 | |
either couldn't or shouldn't do uh you,2020.019,5.64 | |
know a forklift right yes a human could,2023.019,4.681 | |
you know unpack a pallet and put,2025.659,3.601 | |
everything in the you know on a new,2027.7,2.88 | |
palette in the truck but it's easier to,2029.26,3.12 | |
just use the forklift to pick up the,2030.58,3.54 | |
pallet as a whole and put it in the,2032.38,4.08 | |
truck for you likewise there's going to,2034.12,4.439 | |
be some things that generative AI it can,2036.46,4.079 | |
just do that a human can't,2038.559,3.661 | |
um or that would take a human way too,2040.539,4.14 | |
long to do so just offload those things,2042.22,4.14 | |
and let the machine do it because it's,2044.679,3.66 | |
already better than the human,2046.36,6.72 | |
faster so generative AI is super super,2048.339,8.701 | |
fast it can read literally like 20 pages,2053.08,6.779 | |
of text in a few seconds no human is,2057.04,5.52 | |
capable of that and then it can use that,2059.859,4.081 | |
text it can either summarize it,2062.56,4.02 | |
brainstorm extend it it can draft a new,2063.94,4.919 | |
version all of this happens really,2066.58,3.839 | |
really fast and of course time is money,2068.859,4.32 | |
right velocity is speed plus Direction,2070.419,4.2 | |
so you got to make sure you're going in,2073.179,4.2 | |
the right direction but basically so,2074.619,4.141 | |
this is another thing that I do when I'm,2077.379,4.081 | |
Consulting is I say what takes the,2078.76,5.28 | |
longest you know so when I consult with,2081.46,4.679 | |
scientists and academics it's like oh,2084.04,5.16 | |
well you know writing grants is takes,2086.139,4.5 | |
forever,2089.2,3.179 | |
um you know Grant proposals Grant,2090.639,4.921 | |
reviews that takes forever or writing,2092.379,4.98 | |
research papers and revising research,2095.56,3.84 | |
papers takes a long time,2097.359,4.98 | |
whatever business activities take the,2099.4,6.06 | |
longest those might be good targets uh,2102.339,5.641 | |
for generative AI uh not related to,2105.46,4.639 | |
generative AI but many many years ago,2107.98,5.52 | |
earlier in my it infrastructure days I,2110.099,5.201 | |
was working at a small uh medical,2113.5,5.28 | |
startup medical software startup and I I,2115.3,5.34 | |
we got a couple new servers and I added,2118.78,3.42 | |
these new servers to their build Network,2120.64,4.5 | |
and so then their software builds took,2122.2,5.639 | |
went from 26 hours to two and they said,2125.14,4.5 | |
well Dave you threw off our entire,2127.839,4.141 | |
workflow because instead of kicking off,2129.64,3.9 | |
a build and then just coming back the,2131.98,3.48 | |
next day we just go to lunch and come,2133.54,4.039 | |
back and now we have to keep working,2135.46,5.22 | |
but again if you can I mean and,2137.579,4.721 | |
Technology can do this you can shorten,2140.68,4.439 | |
things by a factor of 10x or 100x I,2142.3,4.319 | |
usually look at the things that can,2145.119,4.201 | |
happen that we can shorten by 100x say,2146.619,4.861 | |
okay what takes the longest here,2149.32,4.56 | |
we've got some lawyer friends right the,2151.48,3.72 | |
you know there's there's a bunch of,2153.88,2.699 | |
active lawsuits out there and I said,2155.2,4.5 | |
okay given that we have generative AI,2156.579,5.161 | |
that can read and summarize you know,2159.7,4.56 | |
hundreds of pages of legal documents,2161.74,5.339 | |
really fast how can I build a tool that,2164.26,4.92 | |
is going to really help you get to the,2167.079,4.561 | |
bottom of these things really quick uh,2169.18,5.22 | |
so again looking looking at ways that,2171.64,5.34 | |
generative AI can speed up really basic,2174.4,5.04 | |
stuff just reading right don't,2176.98,4.619 | |
underestimate the power of speeding up,2179.44,4.08 | |
really basic tasks,2181.599,5.281 | |
uh so another example is that I have I,2183.52,6.12 | |
I've built my own tools that I use for,2186.88,5.58 | |
uh for helping revise my writing,2189.64,4.92 | |
um Chad GPT still sucks at writing by,2192.46,3.24 | |
the way,2194.56,2.94 | |
um but so I don't use it for writing I,2195.7,3.84 | |
use it for revisions and providing,2197.5,3.3 | |
feedback because that's something that,2199.54,2.88 | |
is really good at it says Hey I just,2200.8,3.24 | |
read this whole chapter let me give you,2202.42,3.419 | |
some feedback on it and it gives really,2204.04,3.36 | |
good feedback,2205.839,3.121 | |
I know because I've paid a professional,2207.4,4.38 | |
editor and it gives me uh feedback at a,2208.96,4.8 | |
very similar level as a professional,2211.78,4.319 | |
editor now that being said it does miss,2213.76,4.92 | |
a lot of stuff but instead of a two,2216.099,4.921 | |
month turnaround I get a 30 minute,2218.68,4.62 | |
turnaround so that's what I mean by look,2221.02,3.96 | |
for things that you can accelerate by a,2223.3,4.68 | |
factor of 10x or 100x so let's see going,2224.98,5.54 | |
from two months or 60 days to 30 minutes,2227.98,6.599 | |
that's you know 60x and then instead of,2230.52,6.4 | |
hours it's minutes like so I mean you're,2234.579,6.121 | |
looking at a 200x speed up by using chat,2236.92,7.86 | |
GPT API for getting uh feedback rather,2240.7,7.32 | |
than using a human so again like you,2244.78,5.819 | |
know those that level of speed up is,2248.02,4.5 | |
absolutely possible with generative AI,2250.599,3.541 | |
look for those and you're going to,2252.52,3.78 | |
really make a a very good difference in,2254.14,3.78 | |
some departments,2256.3,4.5 | |
and then finally cheaper uh here's the,2257.92,6.3 | |
thing is that a lot of stuff costs a lot,2260.8,5.4 | |
and human labor is one of the most,2264.22,5.46 | |
expensive parts of of uh many businesses,2266.2,6.06 | |
and the best expense is no expense so,2269.68,5.399 | |
what I mean by that is unfortunately we,2272.26,4.62 | |
are seeing where like many many,2275.079,4.141 | |
companies are doing huge huge layoffs as,2276.88,4.199 | |
they pivot and I don't know if those,2279.22,3.48 | |
jobs are ever coming back but that kind,2281.079,3.181 | |
of goes outside of the scope of this,2282.7,3.419 | |
video but,2284.26,4.02 | |
what I mean by that is that if there is,2286.119,4.561 | |
something that you can do cheaper with,2288.28,6.12 | |
AI than paying a human or you can,2290.68,5.82 | |
Empower a human to do some to do the,2294.4,4.5 | |
same task faster and therefore cheaper,2296.5,4.68 | |
that is kind of the direction to look at,2298.9,4.62 | |
so where can you lower costs with,2301.18,4.38 | |
generative AI this one is actually a lot,2303.52,3.42 | |
harder to do,2305.56,4.86 | |
because uh again it's it's better at,2306.94,5.1 | |
some things and it's certainly faster at,2310.42,3.54 | |
some things uh but you still need human,2312.04,4.02 | |
oversight you still need human uh you,2313.96,4.26 | |
know guess and check or whatever uh you,2316.06,5.039 | |
need experts supervising these systems,2318.22,5.04 | |
um but with that being said there are,2321.099,3.961 | |
ways that you can probably lower costs,2323.26,4.68 | |
because again chat gbt has a 20 a month,2325.06,4.92 | |
assistant it can do the work of,2327.94,3.78 | |
literally like thousands of dollars,2329.98,3.84 | |
worth of cognitive labor for you per,2331.72,3.06 | |
month,2333.82,2.34 | |
um so that you don't need to hire like a,2334.78,2.94 | |
research assistant or that sort of thing,2336.16,3.54 | |
which then frees up the you know,2337.72,4.92 | |
cognitive labor of other humans but,2339.7,5.399 | |
again look at where it's cheaper uh,2342.64,4.439 | |
generative AI in the form of image,2345.099,3.841 | |
generators much much cheaper than,2347.079,3.361 | |
graphic artists which is really really,2348.94,3.419 | |
unfortunate for the graphics artists out,2350.44,3.72 | |
there and I don't have a solution but,2352.359,3.961 | |
from a business perspective this says,2354.16,4.56 | |
like okay hey we can get a lot more art,2356.32,5.759 | |
done uh much faster and much cheaper uh,2358.72,5.34 | |
than than with the humans you still,2362.079,3.661 | |
often need humans to do touch-ups and,2364.06,3.36 | |
and that sort of thing and to even Drive,2365.74,4.379 | |
the machines but again much cheaper,2367.42,4.199 | |
because it is faster,2370.119,4.74 | |
you can also use it to do a SWOT,2371.619,4.98 | |
analyzes and that sort of thing again,2374.859,2.961 | |
that goes back to the cognitive,2376.599,3.841 | |
reasoning where you can use generative,2377.82,5.08 | |
AI to help look at and control those,2380.44,4.44 | |
outflows,2382.9,4.92 | |
and then so this is uh we're closing in,2384.88,5.28 | |
on the end of the video one thing that I,2387.82,3.66 | |
really want to point out particularly,2390.16,3.6 | |
for the for the managers middle managers,2391.48,5.28 | |
and the executives watching is that you,2393.76,4.98 | |
need to be using generative AI to,2396.76,3.839 | |
augment yourself,2398.74,3.839 | |
um whether this is using tools like chat,2400.599,4.321 | |
GPT to develop your sense of empathy and,2402.579,6.0 | |
to be more diplomatic uh to even be more,2404.92,5.699 | |
vulnerable to learn to be more,2408.579,4.861 | |
vulnerable with your customers your,2410.619,5.521 | |
employees uh your star players that sort,2413.44,5.58 | |
of thing you use that and you will,2416.14,5.219 | |
develop better relationships and and,2419.02,4.86 | |
you'll develop more trust and have a lot,2421.359,5.041 | |
more of that intangible benefit of,2423.88,5.76 | |
working with your employees and but,2426.4,5.219 | |
employees inside the company and then,2429.64,3.38 | |
customers outside the company,2431.619,4.98 | |
introspection so this is one of the most,2433.02,6.52 | |
unsung things that all leaders need to,2436.599,6.781 | |
do is using these tools whether it's,2439.54,5.88 | |
chat GPT or the reflective journaling,2443.38,4.52 | |
tool that I built or anything else,2445.42,5.04 | |
understanding how your own mind works,2447.9,5.14 | |
and What Makes You tick is absolutely,2450.46,4.92 | |
critical to being a good leader because,2453.04,3.78 | |
if you're doing things for the wrong,2455.38,3.479 | |
reason you need to be aware of that so,2456.82,3.36 | |
that you can fix that and make sure that,2458.859,2.901 | |
you're doing things for the right reason,2460.18,3.6 | |
introspection and the way that,2461.76,4.359 | |
generative AI can help with that is,2463.78,4.799 | |
really really good and then finally,2466.119,4.5 | |
smarter if you're not using generative,2468.579,4.861 | |
AI to learn more faster to make sure,2470.619,4.801 | |
that you have good understanding you,2473.44,4.98 | |
will be left behind and so this is,2475.42,4.86 | |
something like I know I'm saying that,2478.42,3.9 | |
like really bluntly but the fact of the,2480.28,3.78 | |
matter is if you're not using generative,2482.32,3.539 | |
AI to make yourself smarter to make,2484.06,4.86 | |
smarter decisions your competition is so,2485.859,5.041 | |
it's that simple,2488.92,4.8 | |
and then lastly ethics and courage,2490.9,4.679 | |
I've mentioned a few times during this,2493.72,3.66 | |
video that in some cases I think,2495.579,4.201 | |
generative AI will inevitably lead to,2497.38,5.3 | |
layoffs job loss and that sort of thing,2499.78,5.16 | |
uh there's going to be really tough,2502.68,5.62 | |
decisions ahead for a lot of people and,2504.94,6.899 | |
the competition May undercut you and you,2508.3,5.0 | |
might feel like you have to compromise,2511.839,4.081 | |
your principles and your values in order,2513.3,5.74 | |
to stay competitive the thing is it's a,2515.92,5.159 | |
foregone conclusion that technology will,2519.04,3.9 | |
will disrupt markets do you remember,2521.079,4.141 | |
borders borders imploded because of,2522.94,4.38 | |
Amazon that's just the name of the game,2525.22,4.5 | |
so if you need to close the doors and,2527.32,4.259 | |
pivot maybe that's what you need to do,2529.72,3.96 | |
and it really sucks because nobody wants,2531.579,3.601 | |
to go out of business nobody wants to,2533.68,3.54 | |
lay off you know hundreds or thousands,2535.18,4.02 | |
of employees because you're going out of,2537.22,4.5 | |
business that being said it's probably,2539.2,4.34 | |
going to happen,2541.72,4.16 | |
now even if you don't go out of business,2543.54,4.84 | |
layoffs and pivots are coming they're,2545.88,4.239 | |
actively happening,2548.38,2.88 | |
um you know like you remember telephone,2550.119,3.361 | |
operators telephone industry didn't go,2551.26,4.8 | |
away telephone operators did and so,2553.48,4.26 | |
going through these pivots is going to,2556.06,4.26 | |
be very difficult and and very painful,2557.74,3.9 | |
for some people,2560.32,3.36 | |
so what I want to leave you with is this,2561.64,4.32 | |
idea that there are three guiding,2563.68,4.14 | |
principles that you should probably keep,2565.96,6.0 | |
in mind as you Embrace generative Ai and,2567.82,7.259 | |
that is Trust dignity and respect so,2571.96,5.22 | |
first make decisions that prioritize,2575.079,5.101 | |
earning and keeping trust once trust is,2577.18,4.86 | |
lost it is very difficult to get back,2580.18,3.24 | |
and if you look at some of the,2582.04,3.12 | |
technology or tech companies out there,2583.42,4.679 | |
that have abused the markets trust abuse,2585.16,4.38 | |
their customer trust by violating,2588.099,4.201 | |
privacy and that sort of stuff I gotta,2589.54,4.2 | |
tell you like I'm never going back to,2592.3,2.819 | |
those platforms I don't know I don't,2593.74,4.26 | |
care what they do and so if you sell,2595.119,5.281 | |
your trust if you if you sell your,2598.0,4.5 | |
trustworthiness in the age of generative,2600.4,4.5 | |
AI you will probably not be able to get,2602.5,5.04 | |
that back because Above All Else a lot,2604.9,4.439 | |
of people are are caring more and more,2607.54,4.319 | |
about privacy and not being abused and,2609.339,5.461 | |
manipulated uh by the corporations that,2611.859,4.98 | |
we engage with if you watch my my,2614.8,3.539 | |
channel you know that I am very very,2616.839,3.661 | |
skeptical of corporations that being,2618.339,4.201 | |
said I also understand that corporations,2620.5,4.14 | |
are here for a reason and they provide,2622.54,5.4 | |
uh critical goods and services to all of,2624.64,4.32 | |
us,2627.94,4.02 | |
so make sure that you put trust front,2628.96,4.92 | |
and center with the way that you deploy,2631.96,4.32 | |
and use uh generative Ai and the,2633.88,4.199 | |
decisions that you make a strategic,2636.28,3.539 | |
decisions you make around generative AI,2638.079,4.801 | |
dignity so this is something that is one,2639.819,5.361 | |
of those more intangible things which is,2642.88,4.62 | |
you know how do you prioritize human,2645.18,4.419 | |
dignity this means,2647.5,3.72 | |
um this means treating people with,2649.599,4.02 | |
respect respecting the Dignity of their,2651.22,5.16 | |
time off respecting the Dignity of their,2653.619,5.46 | |
competence Their Fear so this is another,2656.38,4.739 | |
thing A lot of people are absolutely,2659.079,4.02 | |
Paralyzed by fear because they don't,2661.119,5.101 | |
know what's coming and as a CEO you,2663.099,4.381 | |
might not know what's coming either but,2666.22,3.119 | |
if you treat people with dignity and you,2667.48,3.48 | |
say you know what I'm not going to,2669.339,4.081 | |
compromise on this value then they're,2670.96,3.54 | |
gonna like you more they're gonna,2673.42,2.699 | |
they're gonna give you more of the,2674.5,3.66 | |
benefit of the doubt and and trust you,2676.119,4.261 | |
to see it through and then finally,2678.16,4.8 | |
respect respect for human rights respect,2680.38,4.56 | |
for privacy respect for the way that,2682.96,4.859 | |
people feel if you prioritize respect as,2684.94,5.879 | |
well as trust and dignity then you know,2687.819,5.101 | |
even if your company implodes or even if,2690.819,4.081 | |
you have to lay off people you're not,2692.92,4.14 | |
gonna you're you're you won't regret it,2694.9,4.98 | |
put it that way you will not ever regret,2697.06,5.64 | |
prioritizing trust dignity and respect,2699.88,5.04 | |
no matter what else happens,2702.7,6.84 | |
and so with this my my caution to all,2704.92,7.08 | |
Executives and leaders out there in the,2709.54,3.9 | |
age of generative AI is do not,2712.0,3.78 | |
compromise your ethics and principles,2713.44,5.04 | |
but also be courageous and make some,2715.78,4.2 | |
bold decisions because if you're not,2718.48,4.92 | |
making bold decisions other people will,2719.98,5.879 | |
so for a quick recap we talked about the,2723.4,3.84 | |
people processes and Tools around,2725.859,4.5 | |
generative AI this is you know,2727.24,4.98 | |
generative AI is just like any other,2730.359,4.161 | |
software it provides new capabilities,2732.22,5.119 | |
new opportunities but also new threats,2734.52,4.78 | |
one of the key ways to look at,2737.339,4.0 | |
generative AI is look at where it's,2739.3,4.2 | |
better faster and or cheaper than humans,2741.339,3.721 | |
this is where you're going to get the,2743.5,2.94 | |
most bang for your buck when you're,2745.06,3.539 | |
either building your own generative AI,2746.44,4.62 | |
tools or you're buying or procuring,2748.599,5.881 | |
generative AI goods and services,2751.06,5.88 | |
uh and then second to last with,2754.48,4.68 | |
self-improvement as an executive leader,2756.94,4.379 | |
if you're not using generative AI to,2759.16,5.04 | |
improve yourself somebody else is namely,2761.319,4.321 | |
your competition and you're gonna get,2764.2,3.899 | |
left behind and then lastly the slide,2765.64,4.14 | |
that we just did was ethics and courage,2768.099,4.441 | |
which means whatever happens whatever's,2769.78,5.46 | |
coming if you if you compromise your,2772.54,4.319 | |
ethics you might regret it but if you,2775.24,3.06 | |
don't compromise your ethics you will,2776.859,3.421 | |
not regret it so choose the path of,2778.3,4.019 | |
least regret this is um actually a,2780.28,4.38 | |
mantra that I live by is what decision,2782.319,4.861 | |
will I regret least in the long run and,2784.66,3.84 | |
then be courageous about it because,2787.18,4.02 | |
again this is a bold time with a,2788.5,4.92 | |
tremendous amount of opportunity but uh,2791.2,4.32 | |
you know fortune favors the Bold so,2793.42,4.14 | |
thanks for watching I hope you got a lot,2795.52,4.26 | |
out of this thank you very much and uh,2797.56,4.32 | |
again reach out on LinkedIn or patreon,2799.78,3.66 | |
and let's talk if you want to if you,2801.88,2.939 | |
want to talk about any of these topics,2803.44,3.98 | |
cheers,2804.819,2.601 | |