Revealed: OpenAI Tells Us ChatGPT's Major Use Cases

Channel: Alex Kantrowitz

Published at: 2025-09-22

YouTube video id: wBPTgrMOoX8

Source: https://www.youtube.com/watch?v=wBPTgrMOoX8

New data from OpenAI AnAnthropic reveals
how people are actually using their chat
bots. Meta's new AI glasses are here.
Are they the road to super intelligence?
Plus, can you actually cancel Jimmy
Kimmel? We'll talk about it on a Big
Technology Podcast Friday edition right
after this. Welcome to Big Technology
Podcast Friday edition, where we break
down the news in our traditional,
coolheaded, and nuanced format. We have
a great show for you today because we
have a boatload of new data from OpenAI
and Anthropic looking at all the
different uses of chatbots. Some data
that may settle some debates we've been
having on the show for quite some time.
We'll also talk about the roll out of
Meta's new AI glasses. Uh some live demo
trouble, but we'll talk about what the
where the product is going. And we'll
also discuss the cancelling or the
pausing of Jimmy Kimmel's show and
really whether cancellation has the same
meaning these days uh given the outlets
available online to continue the
conversation. It's going to be a great
show. Joining us as always on Friday is
Ran John Roy of Margins. Ranjan, great
to see you.
>> Good to see you. I can't believe it.
Open AI just lifted your blog post.
Three ways people are using generative
AI. We're going to get into it but
>> we're going to talk about it. Yeah. Go
ahead.
>> Yeah. I mean, when I was reading that
research, oh man, I was like, did did uh
GPT5 just ingest the big technology blog
and come up with here's a here's how we
want to structure this?
>> I have to be honest, I read that and I
was like, that is the three faces of
generative AI post just coming out in
open AI research. So, I'm so excited to
be able to talk about the data because
we finally do have uh some real data and
let's get right into it. So, OpenAI
releases this post this week how people
are using chat GPT. Uh, I of course was
like, well, what's going on with the
companionship side of things and we
actually have real data on that as well.
But here is here's directly from the
post. Uh, it says, uh, chat GPT consumer
usage is largely about getting everyday
tasks done. Threearters of conversations
focus on practical guidance, seeking
information, and writing with writing
being the most common work task. While
coding and self-expression remain remain
niche activities, practical guidance,
seeking information and writing are the
three most common topics and
collectively account for nearly 80% of
all conversations. Practical guidance is
the most common use case and includes
activities like tutoring and teaching,
how-to advice about a variety of topics,
and creative ideiation. So, Rajan, I
just want to turn this over to you. Um,
what do you think about the fact that
these are the most common use cases of
chat GPT? And I'll just add on another
question because why not? When I saw
this, I said, you know what, chat, like
chat with generative AI and search have
always felt like completely different
things to me. And maybe this is some
evidence for like this bigger overriding
question of is Google Toast why Google
is seeing the success it is and why
generative AI might not be the direct
threat it was portrayed as for so long.
Wait, hold on. I'm going to push back on
that. How do you see practical guidance
in seeking information is distinct from
traditional search? like a wouldn't that
be the main or at least a a very large
component of overall traditional search?
>> Okay, good point. Seeking information
definitely is search, right? That is
without a doubt search. But the fact
that practical guidance is number one
like I think there was a um a
misconception that AI was going to be um
just like search redone, a refactoring
of search. And it's probably led by
publishers who are like, where's our
traffic going? But the fact that
practical guidance is number one, where
you can go to, and we're going to talk
about it more, but you can go to chat
GPT and say, I'm facing this situation.
What should I do? Whether that's a work
situation, a health situation, a
relationship situation, uh getting fit,
which we're going to talk about. That to
me is a brand new use case. You just
have not had that online at all.
>> No, no, no, no. If you think back there
was I remember in the like 2000s or
early 2010s media startup days there was
entire there's one howcast uh where it
was just all short form video that would
just ex do how-to videos like how to uh
do basically small elements of practical
guidance. So, so there was this was a
core component especially like very SEO
optimized traffic. Again, every um
publisher that was trying to like pull
in traffic, it was around these how-to
questions. So, so I think I would still
>> Yeah,
>> that's exactly the point I'm making
here, which is that this has been a the
belief that that AI is the natural
inheritor to search has been driven in
part by people who work in publishing,
which which is a place that the
narratives tend to come out of
>> who've said we've done this before and
this is search uh and this is taking
search. The difference is you had to
create all that content on how cast. Um
this is a technological solution to
practical guidance and that's emergent.
That's something that we haven't had
previously. It's brand new. And so to
me, this idea that it's a technology
replacement to search comes second uh to
this idea that this is a brand new
technology layer uh on the practical
guidance front. And in fact, I would say
maybe this is controversial, but I think
this does a better job than those Hcast
and e-How I and no offense to anybody
who's worked on those websites before.
They are a creation of SEO. They would
not exist without Google and they're not
very good. Like I don't think anyone you
know going giving their their um you
know coming back to a college giving the
commencement address somebody super
successful you know says hey I want to
just thank the people at e-how you know
I had all these problems and I figured
out everything on e-how no it was just
like it could get you maybe twothirds of
the way to fixing your sink
>> when I look back at all the success I
have achieved in my life I would like to
thank eh how
cast Ask Jeeves and all the above. Yeah,
I think
>> students students you you may be asking
how, but you should be asking eh how
>> um we're dating ourselves for any uh Gen
Z listeners here, but there was a time
that these websites were uh were large
and dominant. Um so, okay. So one thing
I'll start to kind of if we dig in
further practical the way they start to
distinguish these things I still found a
little difficult or problematic again
what is the difference between practical
guidance and seeking information like
there's going to be a lot of blurry
lines in there writing I think is fairly
fairly uh consistent but to me they had
gone another level deeper so they had
those as one categorization but then to
me going back to your blog post they had
three patterns within messages. Asking,
doing, and expressing. About half of
messages, 49% are asking. Um, and then
expressing, doing is 40%, expressing is
11%. These fit very neatly into our
discussions, your blog post, thinking,
doing, and companionship. Um, so I think
they start, it's kind of amazing that
they're starting to see the world in the
way that we've been talking about this
thinking, doing companionship or asking,
doing, and expressing. The only thing
for me I want to start kind of like I
want to hear your thoughts on the way
they define doing is not how I define
it. They define doing as getting help in
doing some kind of process like getting
help drafting text or planning or
programming. The way I've been
discussing this is to me doing is
actually going out and doing it. I had a
complaint with Delta about luggage a few
weeks ago and I had chatbt write the
email and tell me where to go, but it
didn't send the email. So, it's almost
there, but it's that to me that is the
real doing. But otherwise, I'm glad
they're starting to look through this uh
in the same way we are.
>> Well, that that's a great point and I'm
going to take it one step further.
Right. So this is these are the tasks
that people are coming to generative AI
or chatpt for. 49% is asking that would
be my thought partner uh category. 11%
is expressing which open AI says is uh
captures uses that are neither doing
asking nor doing usually involves
personal affection or exploration and
play. To me that's companion obviously.
Now doing should be agent, right? And
you're saying that their categorization
of doing is actually much broader than
you would use for agent. And I would
agree. Uh it seems like some of these
things like using the uh chatbot for
drafting text, let's just go with your
definition. If it's not calling a tool,
it's just like normal LLM behavior. So,
I think that what they're doing in this
study is expanding their agent category
to encompass more than it really should.
And I wonder what's why they're doing
that. I think I have an answer. I think
what we're seeing in the numbers here is
the vast majority of activity within
chat GPT is really thought partner. Uh
it is a thought partner tool and the
agentic stuff of course is super early.
Um, but they didn't want maybe the
numbers, and this is maybe somewhat
conspiratorial, but I stand by it. Uh,
they didn't want the numbers to show how
deeply into thought partner they are
when they just released a product GPT5,
not only released, tuned their product
to the agent use case. To me, this
explains why there's been so much uh
uproar and disappointment and disconnect
between the people who've been using the
previous models and the people using
GPT5. OpenAI had a thought partner pro
uh product and they turned it into an
agent product which is by their
definition a minority of the use uses
and as you look a level deeper a very
small percentage of the uses and that to
me is I think it's a problem and that's
why I think these companies are going to
have to have clarity over which one
they're going for. You cannot really do
all three. Maybe you can do it in the
same product but you definitely need a
switcher.
>> Yep. Okay. And I I like this. I don't
think that's even conspiratorial.
Actually, I will say I do wonder like
because this was presented as a research
paper, you know, in as a PDF in that
Times New Roman font and like it just
looks like a research paper. You assume
there's no marketing hand kind of
overseeing it. Whereas if this was
produced as a slick PDF like a from a
Google, you would just be like, "Okay,
whatever. This isn't real." like or it's
just marketing cuz and I I I actually do
wonder how that gets like politically
divided within an organization like
OpenAI because all of the findings
really do help push their business case
like there's uh again like as you said
giving that looser definition of agentic
with doing makes it seem like they have
very distributed use cases and they are
moving towards this world of agent but
where in reality as we're discussing
they're not they also have nearly half
of messages come from users aged 18 to
25. So guess what we're capturing the
younger demographic. They have uh
demographic gaps are shrinking that now
women are using it as much as men. Um
which I kind of found interesting that
>> or the lead finding in this in this
study.
>> Yeah. And but they they were very clear
that it was like with using
traditionally feminine gendered names
like they were like we don't actually
have data on you don't worry but we're
using just
>> but we're going to tell you exactly how
people are using our product down to the
man like tiny little percentage. Okay.
>> Exactly. Exactly. And then they have
like uh emo they said emotional
companionship are rare. They said only
1.9% of messages are about uh
relationships and point4% roleplay.
That's uh that's the companionship we're
talking about. But you know, like
>> that's one one level deeper than
companionship, let me tell you.
>> One asking for help, the other getting
into the role play. Um but but but
overall like everything Oh, they even
had like geographical
uh like lower income countries are
starting to use this more. So it's
positioning it as this beautiful
democratizing force like bridging the
wealth gap between nations. So, so
overall like the more I was looking at
it, I'm like the greatest form of
content marketing you can do is make it
look like a research paper and then
suddenly it just adds so much
credibility to it.
>> Yeah. Look, I have no doubt that
researchers worked on this. Um, but we
cannot It's just one of those things
when a company releases research, you
have to look at it with a not I I don't
think you you don't believe anything you
read. You just have to like read between
the lines a little bit and you can you
can get some good data. Like I'm sure
that part of this does really does
reflect um the way that that chat GPT is
being used. Like the idea to me that
most of it is this thought partner use
case that makes total sense to me. Um it
it's totally tracks with what I've been
thinking about uh without about the bot.
But you do have to say there are certain
narratives that they want out there. And
you're right. If you put it out as this
research paper, it does do a better job
of advancing those narratives than a
thou does protest too much blog post
saying only 1.9% of chatbt messages are
in the topics of relationship of
relationships and personal reflection.
And in fact, this study I initially uh
you know found out about this study
because I think one of our listeners
tagged me on Twitter and said, "Oh, hey,
look, only 1.9%
are companions." And I had been sort of
beating the drum saying that companion
uh you know is is one of the leading use
cases. Maybe I even said it was the
number one use case. I might have to
revise that after this show. Well, it
was a HBR article that seemed to suggest
that and that HBR article was also cited
in this OpenAI study. So I think they
give some credence to it even though
they found something completely
different. But I would say after reading
the articles that have come out about
OpenAI and listening to the podcasts
about OpenAI and how people are building
relationships with Chad GPT and that's
unhealthy. Uh I if let's just say this,
if I was a researcher at OpenAI, I would
do whatever I could and maybe this is me
getting it completely wrong. I'm I'm
open to that. Um I would do whatever I
could to minimize that use case. I would
say there's only a small PE percentage
of people that are falling in love with
ChatGpt. And by the way, let's just do
our our building relationships with the
with Chat GPT. And let's just do the
back of the envelope math here because
we have 700 million uh users of Chat
GPT. And so uh this would say 1.9% of
people have some sort of relationship
with them, I think. So that would still
leave us with 13.3 million people weekly
having a relationship style conversation
with Chat GPT. Obviously, it's not the
overriding use case, but that's still
ton of people.
>> Oh, actually befriended the bot.
>> If you look at it, it's 18 billion
messages per week. Um, so I just did
that's 35 million uh relationship
messages right there per week.
>> Can we can we back the envelope figure
out how much it's costing Open AI to
serve these online boyfriend and
girlfriends?
>> Well, actually, they're going to burn a
hundred. Go ahead. in the world of It's
also funny because like in the world of
GPT5
where you could have a simple
non-aggentic answer that can keep things
going instead it's going to go do some
like multi-dimensional multi-layered
thinking that's just burning tokens just
to say that's a great answer you are so
smart
well you know it is interesting because
you do get I really I mean maybe I
should just do a test but I really
wonder how this agentic you know we
talked last week about how GPT5 always
asks like do you want me to do this for
you like right your friend at the
barbecue where it's like uh where you
said hey are you flirting with with uh
Chad GPT and it's like hey I can be more
flirty if you want like I do wonder when
you get deep into those role plays like
what Chad GPT is actually suggesting as
terms of the next step
>> I someone for research go out I just
can't I I just
that that scares me too actually that's
I feel like Yeah, we need we need we
need that story out there of someone who
actually goes down that that rabbit
hole. Alex,
>> I'm not going to say
>> I'm nominating you.
>> I'm not going to say I'm going to do it,
but I'm also not going to say I'm not
going to do it. If I don't have a story
for next week, I may have to desperately
start to try to uh go into a roleplay
with Chad GPT and see what happens. Uh
God God help us all. There was also
another interesting speaking of stats
here um another interesting stat they
say four 4.2% 2% of chat tpt messages
are related to computer programming
compared to 33% of workrelated uh clawed
conversations. It's interesting how they
have that cl that um qualifier
workrelated cloud conversations. But
even still, I think what they're saying
with this is, hey, uh, we've really been
working on coding and we have a lot of
opportunity here uh, on the coding front
given how uh, how intensely how intense
claude is used for coding and and uh,
sort of how it's still emerging as a use
case for us. Yeah, I I agree that that
one jumped out at me and and we have
been talking about this all for months
now and I I think both of us have agreed
that like Claude's essentially pivot
towards coding as like a core use case
in terms of monetization has actually
been a seemingly successful one as
they've been just like ripping through
revenue growth. But but yeah, I it felt
like as the new codeex product came out,
like this is another one like, hey
everybody, we have plenty of opportunity
here and Claude's already a little bit
saturated. So So it definitely again
this one felt like one of those very uh
very convenient
uh statistics that was that was put out
there. I I I'll say though actually no
this one and it and I wasn't surprised
by it either because again I feel
everyone I speak with chat GPT is not
the default for coding assistance and
coding help. So this made sense.
>> That's right. Okay. Let's talk a little
bit about work versus not work. 30% of
consumer usage is workrelated uh and
approximately 70% is nonwork. That is
interesting to me.
>> Yeah. I think but again how this stuff
gets defined I think is uh is difficult
but but also I think we've said before
that they have to win consumer and they
position themselves as they're going to
win consumer and they they are leading
in consumer. So I think this also still
kind of helps drive that narrative too
that and and apparently that it's up
from 53 to 70% that is nonwork usage.
So, they're showing that this is growing
in importance. Everyday people are using
chat GPT, which I again flirting with it
while it's telling you how to grill on a
Labor Day weekend. Like that's I'm going
to call that nonwork usage. Um so, so
overall, I think that one seems clear to
me.
>> That's right. And um there's there's one
last part here which I think is worth
talking about uh because they they
particularly call out this decision
support side of things. Um and I think
it's very interesting. So they say a key
way that value is created is through
decision support. Chat GPT helps improve
judgment and productivity especially in
knowledge inensive jobs. And as people
discover these and other benefits, usage
depends with uh with user cohorts
increasing their activity over time
through improved models and new use case
discovery. Uh a couple things on this.
First of all, they are saying that this
is a way that chatbot uh uh productivity
is actually uh you can't see it in GDP
numbers because it's not like a clear
activity, but they say that this is
actually already making a difference uh
in terms of economic uh impact. Uh the
other side of it is the fact that
they're highlighting this and practical
guidance um so prominently in this study
just suggests to me that I don't think
we have fully grasped the level with
which people already in the year 2025
3 years nearly 3 years after the release
of Chacht
we trust these things we do we trust it
for guidance in our personal
relationships in our work world, uh,
everything from our health to, um, how
to, you know, write an email inviting
people to a party. Uh, for instance, I
just wrote an email, uh, inviting people
to a party. And, uh, I just
screenshotted it and dropped it in chat
GPT and said, "How's my subject line?"
And it like suggested three different
subject lines. I'm not really the I I
don't have a lot of experience inviting
people to parties, but Chat GPT has a
lot of that baked in. And I was like,
"You know what? your subject line is
better than mine. Copy paste and away it
goes. And the RCPs are flowing in. So,
um, just the amount of trust people have
in these bots is unbelievable already.
>> Is that is that doing right there or is
that thinking?
>> Well, you're the bot's not doing it. I
mean, maybe the bot is, but I would put
that in thought partner.
>> Yeah. Yeah. So, that's where But I would
I'm guessing that would have been that
would have been categorized as doing in
the in the OpenAI context. I also that
is an interesting point around how GDP
and traditional economic metrics aren't
capturing any of this. um because
they'll they're only going to capture I
guess improves improvements in output
but all this and and this happened I
feel like in the early digital days
where there was all this discussion
around the time spent
uh posting on Facebook or posting on
Twitter or all these kind of things
actually created very little economic
value and it was more and it was more
where you know people were just spending
time and energy and I think that's
actually going to be an interesting as
people just spend more and more and more
time with AI chat like what that
actually means for overall measurement
of economic activity because none of it
is going to be captured.
>> Yeah.
>> Other than your other than your better
converting subject lines to get people
to the party that'll be captured
certainly. But
>> that will be captured. Yes. And drinks
served. But um what do you think about
this idea that people trust it already?
I mean for all the it and and is it
trustworthy?
>> Go ahead.
>> Okay, two separate questions there. I
think do people trust it? Uh yes. I mean
undoubtedly everyone I think that we
definitely have crossed the the the
inflection point of all my normie
friends using chat GBT regularly or
using some kind of chatbot Gemini Claude
whatever for in day-to-day personal life
and just asking it questions. Is it
trustworthy?
I think I don't know. It's still tough
because like hallucination I still very
clearly get hallucinations on factual
information if if the quality of
information around the query is not
good. So it's certainly not trustworthy
around more niche topics or things where
there's kind of like conflicting
information out there because it
shouldn't be trustworthy. That's not how
an LLM is meant to work. Um, so I think
that could be more of a problem going
forward, but still right now everyone's
in the kind of honeymoon phase. What
about you?
>> That's right. I I am I do think it is
very interest. I can't tell you whether
it's good or bad that so many people
trust it. I don't know yet. I don't
think we have enough data. Um, I have
found it to be trustworthy to a certain
extent. Um, for instance, like in
situations where I've been sick, I've
said, "Give me a, you know, a dayto-day
of where my health is going to be on
Monday, Tuesday, Wednesday, Thursday of
this week." And it's been pretty
accurate and it's been able to sort of
help me figure out what to do, uh, how
to how to plan things, you know, when to
take it easy, when not to. Um, and
that's just one small use case. I found
it to be obviously it hallucinates but
it gets stuff right so often that I
found myself trusting it more and more
and it also goes back to this Matthew
Prince thing uh the CEO of Cloudflare
that came on a couple weeks ago and he
said people used to click out to the
links now they trust the bot so much
they don't even care about the footnotes
anymore. I think we're getting to that
point.
>> Yeah. And I I think also and is the web
dead? Is the web in secular decline? for
uh any longtime listener will know this
is something we've debated regularly. I
think just so much of the web actually
became less trustworthy or just overly
SEO optimized or just unusable that
people are actually just excited that
there's something that cleanly gives you
an answer that at least seemingly or is
fairly accurate. So people are just
excited about that. But
>> yes, but uh the the one tragedy here is
that E-How is relied upon less. So I
expect us to produce a generation of
less capable leaders because of this.
>> Let's bring it back.
>> e-how.com
e-how maybe they should merge with jet
gpt. By the way, one last thing um the
usage numbers which they included in
this report um with some really like
kind of key data points in terms of how
people have used chat GPT it is
fascinating. So we talked on the show a
lot in 2023 2024 about like where's the
growth of chat GPT is Chat GPT
flatlining and you actually see that uh
getting into mid 2024 it had not reached
200 million users uh after people said
it you know hit 100 million early 2023
which you look at the data and it's not
quite clear that it did at least weekly
users but then around mid 2024 you see a
spike so it crosses 200 million users In
around July 2024, it goes uh up to 300
million in around January 2024.
This year, it's on track to hit 800
million weekly users by the end of the
year. Maybe even more, maybe 900
million. That is astonishing growth. And
I have never seen this in any product
ever before. And to be honest, not even
close.
>> No, I I I agree. And again, you see it I
think like with a lot of technology when
it really is in front of you daytoday in
front of not just early adopters but you
know even the iPhone I remember I stood
in line for the first iPhone and I was
made fun of it my work by people are
like why are you standing in line for a
phone but it was a solid like five or
six years before you saw it just
ubiquitous and uh this has become a lot
lot faster where it's just fully
immersed in everyone. My parents, uh,
pop culture, it's just out there. So,
I'll uh I'll give and it's a testament
to just how revolutionary the actual
products are.
>> That's right. Yeah, it's in South Park.
So,
>> it's in South Park, which was canceled
this week.
>> Yeah, we
>> South Park was canceled this. No, no,
no. They removed the they did not air
their new episode and they said that uh
they said that
>> uh that they did not have it done, which
is just
>> odd.
>> That never happens.
>> Yes. Okay.
>> No, I mean, yeah. Anyways,
>> we'll talk we'll talk about more media
weirdness uh after the break. And if you
enjoy data about the chatbots, we have
more coming your way as we get into the
numbers that Anthropic shared this week.
It seems like it's AI usage data week
here on Big Technology Podcast Friday
edition. And we'll do more of it right
after this. And we're back here on Big
Technology Podcast Friday edition. All
right, so first half we talked all about
this fascinating data from OpenAI. I
feel like that could have been an entire
show on its own. Really fascinating
stuff. I do love it when these companies
publish the data. Uh even if we have to
question some of their motives, it's
always very interesting to get a picture
of where things are going and we have
more data from Anthropic. So Anthropic
released its economic index report this
week. They say uneven geographic and
enterprise AI adoption. Uh a couple
interesting stats right off the top. In
the US alone, Anthropic says 40% of
employers report using AI at work, up
from 20% uh in 2023, two years ago. Uh
such rapid adoption reflects how useful
this technology already is for a wide
range of application. Its deployability
on existing digital infrastructure and
its ease of use by typing and speaking
without specialized training. Rapid
improvement of frontier AI likely
reinforces fast adoption along each of
these dimensions. Here's a little
history lesson they give us.
Historically, new technologies took
decades to reach widespread adoptions.
Adoption electricity took 30 years. The
first mass market personal computers
reached early adopters in 81, but did
not reach the majority of US homes for
another 20 years. Even the rapidly
adopted internet took around 5 years to
hit adoption rate that AI has reached in
just two years. So Raj, I'm turning it
to you. Uh what are we supposed to what
what should we make of the rapid
adoption of AI? Is it just a nature of
the fact that it is so easy to use as
anthropic is suggesting here? And does
that mean we're we're like the um you
know we should be more skeptical because
it's been adopted so widespread and
we're still trying to sort of find out
what the ROI is or is it just a bull
bullish sign because there's been such
enthusiastic adoption that uh it's just
going to get crazier from here.
>> Yeah. Yeah. It's the right question
because like the moment you're like this
is bigger than electricity and I think
wasn't it Sundar who said like it's the
biggest thing since fire.
>> I was in the room for that. I was in the
room. I was talking on uh an NBC show. I
think Car Swisser was hosting it and he
said it it was bigger AI would be bigger
than electricity and bigger than fire
and everyone's like so this Sundar guy
he's on drugs but it turns out that you
know if you said it now we'd have to at
least take that seriously
>> bigger than fire I think. Uh so yeah the
speed of it I I really was trying to
think through like what would be the
correlary is it electricity took because
electricity is a very like heavy
infrastructure type of innovation so to
actually diffuse it across houses or
just people is going to take longer. I
don't know like uh if you think about
viral apps, could Tik Tok be somewhat
comparable even in terms of scale and
speed? Um yeah, I I don't know. It
sounds exciting. It sounds big that to
me still though this like millennia like
looking in the context of like
innovations over centuries and
millennia, I'm still not there. I'm
like, you know what? People are people
are flirting with Chad GPT a lot more,
but is it that revolutionary just yet?
We still have to wait and see. I
>> I'll say this. If I was running uh
again, these studies are marketing in
some way. They're also helpful and
informative, but we can't separate the
two. If I was running uh the publication
of one of these at a company uh whose
product was taking off this way, would I
compare it to electricity and fire? I
mean, hell yeah. that my job was
marketing. I think I would um if I'm but
I'm looking at it now. Am I going to say
it as an impartial observer that it is?
No, I'm not. But I do think that the the
um fast adoption rate is definitely
notable and I think we'll probably have
answers about what this technology will
actually do sooner rather than later
given the investment and given the
attention and we've seen we so so
Anthropic didn't just give us some of
this data about enterprise use. By the
way, I think 43 40% of enterprise use.
Maybe that's on API side of things. Uh
we know that many more people are using
this chatgpt within their organizations.
But here's like some interesting data on
what cloud is being used for. So coding
is still number one use case at 36%.
But educational tasks have surged from
9.3% to 12.4% and scientific tasks are
now uh 6 they went from 6.3% to 7.2%.
2%. This is an interesting thing for
you, Ranjan. Users are entrusting Claude
with more autonomy. Directive
conversations where users delegate
complex delegate complete task to Claude
have jumped from 29 27% to 39%.
Uh we're seeing an increased uh program
uh creation and coding and a reduction
in debugging suggesting that users might
be able to achieve more of their goals
in a single exchange. So Claude being
used for uh education, science and more
complete tests. Go do this claude and
then people trust it.
>> Yeah. Actually and again as we've
discussed those all very neatly fit into
anthropic strategic objectives. So as we
get into these it's like we know
anthropic is pushing Claude into more
coding more kind of like educational
scientific use cases.
very conveniently. I will say I was just
looking they presented this research not
in that academic PDF format but actually
as a blog post. So my one my one uh call
out to anthropic just make it a PDF that
looks like an ep academic paper and uh
we'll we'll trust it more. But but
overall it's all of this stuff I think
is interesting. I think it's it feels
correct. Again, I I don't know anyone
who's using Claude that much in terms of
the thought partner side of it other
than my my co-host of margins, uh, John.
He's been using he's been he's like a
claude head for thought partnership
through and through, but but overall,
yeah, I don't know anyone else who's
who's using it. So, this this all seems
to add up. I unfortunately have uh some
news to report to you here which is that
there is a big group that seems to be
using it for thought partnership and
that is the US government. Uh the city
that or region that had the most clawed
adoption per capita is Washington DC.
So US government seems like it's being
run by
>> cloth.
Um, so
>> which frankly might be an improvement.
>> I don't I don't know what that says
about Claude right now given
>> cancelled Jimmy Kimmel.
>> Yeah, Claude uh Claude is testifying
gonna be pulled up in front of Congress
and has to uh answer some questions.
Yeah. I wonder though, see this is where
the data gets kind of just questionable
like the way they define per capita
usage. I had actually looked, they had a
formula around like total usage relative
to overall population and uh I mean if
you just have a couple of big federal
contracts that are using claude and DC
is a smaller population density area
relative to like uh you know any most
other cities that would skew that data
completely. So I don't know that one uh
that one seems not the most uh easy to
understand. Let's get to one headline
that I think was pretty interesting uh
that sort of got the most attention here
and that is automation tasks are
surpassing augmentation for claude. uh
they cl sorry anthropic says 77% of
business uses involve automation uh
compared to about 50% of claude users
and you look at the chart that they put
put together and automation was actually
a um was actually much less 41%
automation compared to 55% augmentation
uh I think this is in their v1 today
automation has has passed uh
augmentation 49% automation 47%
augmentation. So this idea that are
people trying to augment or automate
work uh the study seems to suggest that
automation is taking priority. However,
I I will just say that often times you
can automate work tasks and then free
somebody else free someone up to do
something else. Is that an augmentation
task? No. It's probably in the data
scene as automation, but it's actually
uh it's actually the the same thing as
an augmentation
if you guys.
>> No, no, no. I I I agree. Like again,
what does what is the definition of
automation here? Because in reality, I'm
assuming it's mostly coding because like
claude has some connectors where that
allow you to do stuff with other
systems. I've tried them. They're not
great right now. So I cannot imagine
that at any kind of scaled usage like
people are building these like complex
agentic workflows using it. So so I I
I'm curious how they define that uh
directive. I think overall as as
listeners can feel like when all these
numbers just so neatly fit into the
existing strategic narrative of these
companies, I have a hard time just, you
know, taking it at full face value.
>> Yeah. that. And by the way, that's what
we're here for. Like, we want to provide
nuance in these conversations. We want
to read it. We want to attack it with um
some perspective as impartial outsiders
that you wouldn't get necessarily from
uh someone who's just trying to push the
company lines. So, I think overall, just
to wrap this up, we both say this is
interesting data, I believe, and uh it
is just a data point, I would say, and
not the beall end all. And you also get
incredible recommendations like put your
findings into an academic style PDF
paper and it will increase the
credibility.
>> I mean the marketing agency of Roy and
Cano is we're we're just doing work.
>> That's one of the core offerings right
there. Make your research look more
credible.
>> Put it in a PDF.
>> Yep. Put it in a PDF.
>> Listen, it's these are non-obvious
things. Okay, we're running out of time.
We have two more things to talk about.
So briefly, Meta has the $799
glasses. We talked about it last week.
week. I don't think we have to go into
it too much detail this week, but we now
know um the the uh the truth here. Very
interesting dichotomy during this Meta
event. Uh reviewers gushing over the
glasses. Meta on stage unable to get
them to work. Now, sometimes that's
because the Wi-Fi in an event space gets
jammed, but it was very interesting to
see this happening. To me, big question
here is that there's going to be a
display on these cameras. Um
let's see. Uh German says that uh over
time Meta might allow uh people to
offload some functionality to their
eyewear that would normally be on their
phone. So could this be a replacement uh
for the phone? The display basically
lets you have a viewfinder. You see what
your your photos are going to look like
before you snap them? There's also live
captions, a live captions features uh
that displays uh spoken words in real
time, including translation, similar to
closed captions on TV. That's cool. You
can message, you can talk on WhatsApp.
There's going to be a music app powered
by Spotify. Instagram will initially
only support direct messages, but Meta
plans to add reals viewing later this
year because that's apparently what you
have to do on your glasses is sit back
and watch a real. Um, I'm I'm kind I'm
sounding like a hater. Uh, I'm
interested to use this technology. We
both like the Ray-B band metas. I just
uh to me the idea of putting a screen in
front of my eyes uh in a world that I'm
already sucked in by computing too much
uh is not very appealing.
>> Okay, so this is where I am incredibly
excited about this. I talked about it
last week, but even from and we'll get
into the failed demo, but overall like
this is exactly what I was hoping for.
So, for reference, I have the Meta Ray-B
bands taking photos, asking
very simple questions to Meta AI,
otherwise it doesn't get them. Walking
around New York. I love it. Uh, I also
actually have snap spectacles, not like
the AR spectacles. I'd been part of the
developer program and had been testing
them. So having an augmented reality
screen, they're like big and bulky and
they're definitely a developer product
as opposed to anything kind of consumer
already. I saw the potential and even
like walking around there's different
games you can play with it. You can have
different information feeds. So it's
still very clunky the snap version, but
still you could see where the the
promise is is. So to me this is going to
happen. This is the direction. I think
the screen again, if you're watching
reels while crossing the street, that's
not good. If you have like a light bit
of information to in the 20° field of
view in the far right of your lens that
shows you a text message so you don't
pull out your phone or stare at it as
you're walking, which everyone in New
York City does, I think that's great. So
I think like that slightly augmented
layer of computing the if it's done well
this could be massive.
>> I remain skeptical but I'm willing to
try them and see what happens. I just
again like I want to get offline more
honestly. No, but come on. This is it.
But there is the reality and there's the
kind of like the hope and and to me and
maybe it's New York City more than other
cities but actually I was just I mean in
Paris and like uh same people are just
they have their phones out as they're
walking around and and I think anything
and I actually loved that and I am not
the biggest fan of Meta or its
leadership for many years but like I
loved Andrew Bosworth talking about this
is about keeping your phone in your
pocket. That's what that's exactly how
I've seen this whole interface of
computing for a long time. And they are
they're winning. They're winning right
now and they're they're moving in the
right direction as well. I say the same
thing about the Apple Watch and I'm sure
you've been with people who you're
having a conversation with and their
wrist buzzes and they look at it and
it's some dumb notification and it ruins
the train of conversation there. So
>> yeah,
>> I will reserve judgment here. Am I
looking forward to using them? Hell yes.
Am I nervous? Also, yes, also.
>> And the social normaly of like if you're
in a conversation with someone and you
see their eyeballs kind of like moving
to the right and not looking at you, are
they are they just like scrolling reels
right there? That's uh it'll add some
interesting dynamics to human
interaction.
I mean, you remember all the parodies of
Google Glass? It's like I'm on a date
and I'm like looking up things to say.
Now, especially, could you imagine
you're like on a date like this? There
will be a parody of this. Someone's on a
date and they have Meta AI turned on and
it's teaching them how to be like smooth
talker.
>> Yeah. I mean, hey, you know what? If it
helps those who cannot talk smoothly, is
that a bad thing, Alex?
>> Yes.
>> If it
>> be yourself. Blow that date. Well, she
wasn't right for you anyway.
>> You become dependent on it. Yep,
>> that's right.
>> Just stay chat GPT.
>> She'll always be there for you.
>> Can you imagine a relationship where
someone's like super smooth in like
outdoor environments, but like the
second you go for a swim, they're just
like a total dud. They can't wear their
glasses
>> on the beach.
>> This is gonna be a movie, I'm telling
you.
>> Yeah. And we'll make it on VO.
>> Yeah. All right. So uh they uh they did
talk about the why the demo failed.
Apparently uh Zuck said, "Hey, Meta
Start live AI activated everybody's uh
AI in the room and effectively DDoS the
servers." All of the servers were al all
the traffic was also rooted to the
company's servers uh in Iceland
apparently. So that's why it broke. All
right, I'm not going to kill them on on
not being able to demo live. Go ahead.
I'm so happy. We talked about this last
week, how there's been no live demos.
Apple destroyed what is a live demo. And
I loved honestly like not trying to give
him too much credit here, but I was like
I like a a slightly wonky demo is
awesome is like that's what the world
wants right now. It's been so long. I
actually think it's going to give a lot
more credibility among developers that
this is real and it's wonky and it
didn't work on stage right now, but
they're at least showing what is real
rather than Apple everyone has lost all
faith in in terms of like what's real
and what's not.
>> True. All right, we got five minutes
left. I want to touch on the Jimmy
Kimmel thing. I wish we had much more
time uh for this segment. Basically,
what happened this morning was you and I
were texting and asking each other
whether we should weigh in on Kimmel.
Um, not to speak for both of us. I think
we both believe that it's ridiculous
that the government uh pushed uh Kimmel
off air, even if it's temporary. Um, you
know, if if you're going to take
yourself serious, if you're going to be
taken seriously, you have to be willing
to, you know, to be confident enough to
take ridicule. Goes for the left and the
right. This is just my perspective. Not
speaking for Anjan here. Um, and when
you start to go after comedians, even if
you don't think they're funny, uh, which
I know the government, US government
doesn't think Jimmy Kimmel is very
funny, you look weak. Uh, and to me, I
think that that is that is, uh, what's
happening here. It's, you've seen
obviously, uh, a lot of people speaking
out from every political angle,
basically saying this is ridiculous. So,
I don't think we have much to add on
that front. I think what is interesting
here is the question of can you be
cancelled in the traditional way in 2025
uh because
let's say you were to leave a network um
like Colbear is going to and maybe
Kimmel will after this uh you still have
the opportunity to form an independent
media agency and you could look at a
list of a long line of people who have
done this um Shane Gillis after he was
canceled by SNL before joining the cast
but also Tucker Carlson, Megan, Megan
Kelly, uh even Bill Maher after he was
canceled, although he ended up back on
HBO. So maybe Bill isn't like the
perfect example, but you could end up uh
basically going independent, not being
dependent on any uh any uh company, and
you could do better uh that way. So, um,
you know, obviously like without getting
too much into the details of like the
the whole thing before, I'm just kind of
curious like what you think this
actually means for Kimmel and whether
this is as impactful as it was
previously.
>> Well, I I think taking the second
question first, is this as impactful as
it would have been in the past? No, I
don't think it is at all. In fact, I
mean that playbook of they cancelled me,
subscribe to my substack is something
that like so many people have perfected
in terms of monetizing so well. And in
reality, like from a distribution
standpoint, late night TV is such a
weird format that the more I like I
hadn't spent a lot of time in a lot like
reminding myself how NextStar and
Sinclair and the whole affiliate model
works with uh these networks. But again,
these no one is watching this stuff live
anymore. So the value comes out in the
clips afterwards. There's still value in
the platform to an extent, but in
reality, Jimmy Kimmel is the brand.
Stephen Cobear is the brand. Like the
late night show to or the Tonight Show,
like these things are no longer the
brand anymore. So to me, like you can
get cancelled and if especially if you
make that a central part of your they're
coming for me in the mainstream media.
Even uh Karen Antia, I'm not sure how to
pronounce it from the Washington Post.
>> Yeah. Like I again like she had an
entire Substack post about like they
canled me, they're trying to silence me,
subscribe to my Substack now. So like
it's become such an almost like
frustrating playbook but a successful
one that if Kimmel wanted to, he could
go allin blue sky, subscribe to my
substack and like be just as
influential. I'd be curious though if
someone like him wants to play that
game. I don't think he does, but I don't
know. What about you?
>> Yeah. No, I think that Well, yeah, I
think he could do it. Obviously, it's
like worked. Conan O'Brien is a great
example of a comedian who wasn't
canceled but left and I think has just
as much influence if not more with his
podcast uh now than and obviously much
more control than he did previously. Um,
but I really think it comes down to like
uh and okay, okay, I'm just this is
again us just like you know there have
been some people who've been like well
he had bad ratings anyway and he wasn't
profitable. I don't think that was the
scenario. U but I think the overriding
story of like if you're somebody in his
position what are you facing now? Um,
and how impactful is this? That that is
uh for us a pertinent question. And
there was this great line on CNN, Jeff
Jarvis, God bless him, was there uh and
he told the panel, "The only good news I
see, and I hate to say this to my
friends at CNN, is that mass media are
dying." Um, and it is sort of this thing
where like if you're an entertainer or
in a position like that, it just today
it makes much more sense to own your
audience. And maybe this idea that you
would have a comedian anchoring a
network's evening coverage is just going
to go away.
>> Yeah, I I think and Jeff has been saying
that since 2020 2010. So
>> he's probably not I mean he's obviously
I don't
seem like the trend is counterfactual to
his argument.
>> I'll give him he got it right. But in in
truth though, he was going viral because
he was on CNN. Like I think this is
where there's still there's power in
platform to have those kind of secondary
I agree like if only 129,000 people are
viewing live between ages of 18 to 45
Jimmy Kimmel. Yeah. But the influence he
has is that actually doesn't denigrate
it at all. It means that we all see
clips on Instagram or Tik Tok and
YouTube all day long. So, but but in
terms of the business model, advertisers
are paying to be shown alongside that
live viewing part of it for the actual
TV show, which is dead. I mean, it's
completely dead. So, so I do think we
it's a good reminder we need to rethink
that overall business model. I think
none of that has anything to do with why
he was suspended and it is terrifying
and it's going to get uglier and darker
I think before it gets better. But but
overall I think uh I'm not I'm not
pouring one out for the uh Sinclair
medias of the world.
>> Right. Okay. I know you have to go. One
last thought and then I'll end this is
um you know the next generation of Jimmy
Kimmels are not going to look like this
generation. like the next comedian uh is
not going to have this aspiration to go
host the Tonight Show or the Late Show
on a television network. They or the
Jimmy Kimmel show. They are going to be
just It shows how quickly we've changed.
They're going to be digital native. They
are going to have a YouTube channel.
They're going to have a podcast. They're
going to be the owners of their content
and not be subject to this stuff. So, um
I think it's unfortunate what happened,
but I also think that it's it's looking
forward the government's power to do
this type of stuff is going to be much
much more limited because they don't
have the ability to like go to a YouTube
and say, "Yank Jimmy Kimmel."
>> Yeah. And just like both of us who own
our own audiences, channels and content.
Yeah. Yeah. But just cancel us so we can
go on tour saying we got cancelled.
Subscribe.
>> All right. Get us in the Discord. Join
the Discord. Join, become a paid
subscriber, join the Discord, revolt,
overthrow Ranjan and I, and we will get
working on the next iteration of e-How.
So, thank you all for listening. Thank
you, Ranjan. Uh, always great to talk to
you, Rajan. Thank you again.
>> See you next week.
>> All right, everybody. Uh, next week, I
believe we're going to have Enon Costa,
the co-founder of Whiz, talking about AI
and cyber security. So, we will see you
next time on Big Technology Podcast.