Loading...
Loading...

Brian Maucere and Beth Lyons open with carryover news tied to Anthropic’s “Department of War” commentary and the online reaction to Sam Altman’s weekend AMA on X. They discuss the “Quit ChatGPT / Quit OpenAI” chatter and how switching incentives and politics can shape AI platform narratives. Later, the conversation shifts to AI authenticity and editing—using Nate Jones as the jumping-off point—touching on uncanny eye-tracking, disclosure expectations, and audience trust. They wrap with a quick scan of smaller developments (e.g., Copilot “Canvas” leak and model-leak buzz like “ChatGPT-V”).
Key Points Discussed
00:00:18 Opening + what’s on deck (Anthropic “Department of War,” Sam Altman response, uncanny valley topic setup)
00:01:26 Sam Altman’s Saturday-night AMA on X and the “switching to Anthropic” zeitgeist
00:16:59 “Quit ChatGPT / Quit OpenAI” movement and Anthropic’s “easy switch” prompt framing
00:19:50 Tim Urban “Wait But Why” reference as a framing/analogy moment
00:30:47 Topic shift: “I do really want to bring this up” → Nate Jones and the AI-editing authenticity debate
00:42:59 Uncanny tools: Descript-style eye tracking / “underlord” editor talk and why it distracts
00:47:44 Responding to “AI witch hunt” comments; broader point about disclosure and audience trust
00:50:17 Quick hits: Microsoft “Copilot Canvas” freeform workspace discussion (and other small items)
00:51:01 “One more thing” before wrap: “ChatGPT-V” leakage chatter and skepticism about leaks
The Daily AI Show Co Hosts: Beth Lyons, Brian Maucere, Karl Yeh
What is going on everybody?
It is March 2nd, 2026.
This is the Daily AI show and welcome.
We've got here here.
Hope y'all had a great weekend.
Beth Lyons is here with me today on Brian Misari.
I think Carl will be popping in the door here,
maybe in a little bit.
And yeah, I think this week we today,
rather we've had some like carry over information
or news that's come over from like Friday, Thursday, Friday.
A lot of it revolving around Anthropics response,
which we did talk about on Friday show with the Department
of War and different parts of US governments
and beef they had there.
As well as we've probably got a lot more news after that
was Open AI, Sam Altman's response,
which we'll definitely get into.
As well as a bunch of other stuff like Uncanny Valley
and with Nate Jones, like that was weird.
So I maybe want to get into that as well.
So lots to talk about on the show today
to kick off our week.
Beth, how are you?
How are you doing?
I am doing OK.
Yeah, I was glued to the story all weekend.
Sam did a random Saturday night, ask me anything on X.
Just like, hey, y'all are talking about the thing
that we're talking about.
So let's talk about it kind of stuff.
It was probably the right move I would imagine.
Right, but those things usually happen on Reddit.
But this conversation was on tick.
Oh, yeah, I just said when you said it,
I didn't even hear you say not read it.
That's why this conversation I think was just happening
so much on X and the zeitgeist of, well,
I guess I'm switching to anthropic, right?
And is it the same deal?
Is it not the same deal?
So that was really interesting to see play out.
Yeah, and I'll be honest with you.
I don't know that I really have my finger on the pulse
of where it did sort of net out.
Here's what I know.
There was a lot of people that were for their own reasons upset
and there was, you saw a little bit of that whole,
what is it, quit GBT?
I can't remember what the term is there.
But people were saying that they were going to leave
publicly posting unlinked in for whatever their intentions were,
getting a couple of clicks or likes or whatever I don't know.
But people, quick to say that they were done with chat GBT,
or I'm sorry, OpenAI, and they were switching all their accounts
to this side of the other.
Some were saying they were using Mistro,
which I hadn't heard come up in a little bit.
I thought that was, like, one person was like,
we're using Mistro, unlike there's a bold choice.
Nothing wrong with Mistro, by the way,
could be a very, very good bottle.
It just was odd to see come out of nowhere,
as well as some other ones on there as well.
So hard to say, I mean, I'll,
I'm going to ping this back to you, Beth.
What I will say just to kind of tee this up is 10 p.m.
on February, I guess, my time, 10 p.m. on February 27th,
which let's see, that is, that would be Friday, right?
Thursday?
20.
Nope, you're right, Thursday, Friday, 27 was Friday.
Wow.
28 days, 28 days, half September, right?
Like, or whatever it is, 30 days, don't listen to me.
So this is like 24 hours before,
I think he does the AMA, right?
Based on what you just told me,
because you said that was on Saturday, correct?
Yeah.
So this is 24 hours before.
And this is also the same day that Anthropic
had submitted their response to the Department of War,
and unshockingly, the government had, you know,
a strong stance on that of, you know, Anthropic is not patriotic
and they were going to label them a security threat,
which was something that just isn't done,
no matter what you think about it,
it does not have historical precedence to it,
like it is being used or threatened to be used now.
So anyway, that was that.
And then so I'm not going to read the whole thing,
but what I, what I pulled out of this,
and then what I'd be curious about,
if you, if you do have any more information on the AMA bath,
is he says, tonight we reach an agreement
with the Department of War to ploy our models
in their classified network.
But what he says right after that is that two things
that were absolute no-goes for us
are the exact two things that Anthropic had.
They said they wouldn't do, which was mass surveillance,
domestic mass surveillance,
and human responsibility for the use of force,
including for autonomous weapon systems.
This is what Dario said in there in his, in his post.
He's like, we're not, we're not bending on that.
We do not trust the AI to be good enough
to have full autonomy over weapons.
We don't believe that AI should be used for that.
So to me, it was just odd,
because Sam calls out and says,
hey, by the way, we have the same two restrictions,
and the Department of War was totally on board with this.
I'm like, were they?
Because, like, were they just have beef with Dario?
And that was that, right?
That had nothing to do with what he was his concessions
or Anthropic's concessions or otherwise.
And I don't really have a dog in this fight, honestly.
This is way above me.
But he did say at the end of it,
which I thought was interesting,
and I'd be curious more people asked about in the AMA
is we are asking the DRW to offer the same terms
to all AI companies, which in our opinion,
we think everyone should be willing to accept.
We have expressed our strong desire
to see things de-escalate away from legal
and governmental actions and towards reasonable agreements.
Now, he did not say Anthropic,
but I think everybody knows that's who he's talking
about in that particular statement.
At least that's the way I would say it.
And so anyway, that's where I'll take it over to you, Beth.
What else do you know in this?
Because I'll be honest with you,
I don't really know how this netted out.
Do you have any extra on this?
So the language is slightly different.
So the language in the initial contract
with that the Defense Department,
AKA the Department of War, just a little aside,
Congress names the department,
so you can't just announce it and change the name,
even though you've changed the logo or whatever,
and everything that's not its official name,
but it is the name that is being searched on
and we are saying that.
So the Pentagon's offer to all vendors,
including Anthropic initially,
was the language along the lines of AI may be used
for all lawful purposes in classified systems.
And that meant lawful without vendor level veto rights
and that's the sticking point.
Open AI said that in their contract
they're basically saying the same thing,
but they're not using the all lawful purposes
without vendor veto rights.
And in fact, when Sam was referencing it,
he said the Department of War says
that they won't put us in that position, right?
We have the ability to say we don't wanna be in that position
and the Department of War said, sure, that's okay.
But the actual language in the contract
is not the same legal enforcing thing
that if you went against Anthropics,
you had violated their terms of the contract
that was eliminated in Sam's.
And Sam was to his point,
I mean, Sam's a smart savvy guy, right?
There's a sweet, sweet deal that is now available
for the taking and if you move quickly,
you could take away from Anthropics
something that Anthropic had, which was sweet
and prevent GROC from getting something that GROC wants, right?
I mean, this is a perfect slide in for Sam,
but what he was also focused on is he thought
that the designating Anthropic as a supply chain risk
which is a formal designation in the defense industry
and it means that all government agencies
cannot do business with that vendor.
Sam was talking a lot about,
that's a really dangerous precedent.
And so we want to vehemently say
that we're disagree with that just from a,
like independent capitalism.
Like, no, no, no, this is like huge overreach
in terms of reaction to this.
So that was also part of why Sam wanted to do the AMA
and be able to get his, his voicing on that out.
And I don't know what I feel about the overall things,
I mean, I think, yes, humans should be involved
in deciding when other humans get killed,
but they are to a certain extent involved
a college friend of mine this weekend said,
they don't know, humans need to be involved
in all fact checking.
And she's adjacent to the insider AI stuff,
but I was thinking, oh, yeah, that's not possible now
because the data's coming in too quickly.
Yeah, right?
You're going to get a summary.
AI fact checking of other AI, yes.
Right, that's, that's about the bullshit best
you could hope for is if you throw enough
back checkers again, something AI or otherwise the,
the chances of it being aeronious or go down, you know,
you put more digital eyeballs on it, you know,
artificial eyeball.
And it doesn't even need to be AI,
like it can be a coded specific thing,
but then you lose the nuance, right?
But I think we've already been in a situation
where in times of like needing to make decisions
on a dime, like spur of the moment,
obviously they thought about, you can decide,
whether you think they thought long enough,
but they thought about the decision to go
and escalate this more than they,
they didn't just do that on Saturday,
but every response after that has to be made
in a much faster time period.
And therefore you're talking about the amount of time
that it takes to process data.
Yeah.
Well, I guess on the, on the way I look at this
is, you know, look for, let's look at it
from the PR side of things.
By the way, Carl just joined us.
Welcome, Carl.
We're just talking about the whole Sam Altman AMA on X,
but also follow up to Twitter.
I'm sorry to anthropic in the Department of War
and the whole thing.
But what I was going to say is, you know,
from a PR standpoint, I've said this,
I've said this a couple of times, like, you know,
look, opening eyes big enough that Sam Altman needs to have,
you know, PR responses or whatever, weirdly,
I'll go against myself and say like, well, this,
it's very hard to have real-time PR.
That's spinning answers in an AMA.
Maybe that was his point.
Maybe I'm wrong and that Sam still does largely answer
as he wants to answer and doesn't really send things
through any sort of spin, zone or anything like that.
I'm not sure, you know, I don't know if people feel like,
oh, well, you know, Sam got on here
and was willing to take the hits and answer the answer,
answer the questions.
You know, okay, you know, whatever the case is.
Also, I will just say for my, this is my perspective on this.
I mean, I certainly can watch the news
and read and, you know, we can have a great discussion
about it.
I think people are sort of maybe blowing
some of this out of proportion from the way I see it.
But, you know, maybe I'm wrong on that.
I just think it's like, people's gut reactions
are interesting when I read them on LinkedIn.
At the same time, people should vote with their wallets
and that is one way to do it.
If you feel like, open AI no longer, you know,
is in line with what you believe is right,
whatever that might be, and you wanna vote with your wallet.
And I mean, yeah, great.
I'm glad people are able to do that, you know,
and feel like they can move on to other models
and they can continue whatever.
Conversely, I'm sure people feel the same way about Anthropic
because they may feel that Anthropic turned its back
or because of the way it's being spun for them, on them,
they may feel that Anthropic turned their back
on the U.S. government.
I mean, there's certainly people who have read articles
who feel like that is what happened, you know,
Anthropic wants to be anti-American.
I am sure that's being fed through there.
I don't believe that, but I'm sure there's a lot of places
where that kind of, you know, spin is out there.
And so I'm sure people who have decided not to be
with Anthropic because of that reason, you know,
vote with your wall to guess.
Additionally, when this decision was made,
there was a built-in six-month grace period.
So actually Anthropics Tech was used on Saturday
in the Iran bombing.
Oh, right, yeah, yes, like MPD coupled that fast, right?
It's so like, you may not use it.
Like the designation was, you must cut off all use
of Anthropics Technology.
And then the very next day,
we are absolutely using Anthropics Technology.
Now that's nuance of governmental policy
and all of those things, but on like the broad strokes
it was like, what just happened?
Yeah, and he thought it's Carl before we move away
from this particular topic.
I mean, you probably all said what needed to be said, right?
I think it was like a lot of it is like gut reaction.
I'm gonna do this and then, you know, whatever you choose
to use, it's what it is.
I think like, I can empathize.
I can definitely understand like where some people
are coming from, where are other people
and Beth, you said it great.
I think Wall Street Journal put that article up
or they were saying, yeah, well, a lot of the, you know,
there wasn't this amount of outrage
when Anthropics signed the deal in 2024
with the deployment.
I was like, they've been doing this for like a couple of years.
So like, and my big point is, I think,
whether Anthropics does this, whether OpenAI does it,
if OpenAI didn't do it, Google could do it.
And then if Google doesn't do it, there's XAI
that would do it, that I would do it, right?
It's not like, and then, you know, for Sam to go out,
do his AMA, you know, take the hits, talk about it,
their teams, you know, aligned.
Do you think Elon Musk gives a crap
about your thoughts and feelings about this?
Does he care about your, if you use X or not?
No, he doesn't.
So I think like, of all, I'd rather have Anthropic
or OpenAI be the ones have these things.
Here's what we do rather than, I'm not gonna do any of that,
I just didn't do what we were gonna do.
And when I tell you anything about it, right?
Because, yeah, it's like, okay, sure.
Because, yeah, even if you, even if you're like,
well, Google could have done it,
Google is such a bigger company.
I mean, Anthropic and, sorry, yeah, Anthropic
and OpenAI for you can make the argument
that they're branching out into other things,
they are AI companies, right?
Like, this is what they do.
Google is so much more meta and X,
I just have really different circumstances there.
Yeah, you're right.
Yeah, I'm much harder to, like, probably figure out
where the lines are for what you can and can do with Google.
You know, it's not as simple as saying,
Gemini is now available to the department
or something, you know, I mean, what do I know?
But yeah, all right.
Well, I mean, do you guys, do you want more point on it?
Yeah, go ahead.
If I want other point, it is real Carl afterwards.
There was a movement associated with quit chat GPT
or quit OpenAI when the news came out
that Greg Brockman was like the top investor or donator
to something, and I don't know what he's donating to,
but he's a big supporter of Trump.
And so that had started already and this capitalized on it
because it already existed.
And also Anthropic put out a very nice, like,
hey, want to switch?
Just give this prompt to your to chat GPT
and it can carry over all your memories, right?
Like, so they set it up so that it was easy.
Was cheeky.
Yeah, was cheeky.
I think though, you know, like, we'll see.
I'll, you know, I'm meeting with clients and stuff.
I doubt from a regular sense.
This will be like, oh, yeah, we're switching.
Yeah.
I don't imagine any of that.
Yeah, like, I just cannot imagine.
Yeah.
And I also think too.
It's like, what's unfortunate is like the politicization,
which is it was inevitable of AI, where it's like, hey,
we're kind of forgetting the point here
that there's a whole bunch of stuff that's happening.
Like, this is a big piece, like, obviously,
with the war and so on.
But like, you know, use cases and some of it,
I feel like, when whatever new model comes out
and whatever new thing comes out, it'll be like, oh,
you know that the articles will come out.
It's like, GBT 5.3 or 4 just came out
and that people are doing this in eight days.
Tears 10 mind-blowing examples.
You know, it'll just go and then we'll see it.
And we're like, yeah, I know.
Like, do you notice, like, if you step back a bit,
the AI bubble is very fickle.
So it's like, it jumps so fast from one to another
to another to another.
It's literally a squirrel with ADHD.
It's like, it's just ADD.
It's like, go, go, go, go.
We're going to look at this.
We're going to look at this.
I'm outraged with this, but not really.
But I'm outraged, but not really.
And it's like, yeah, it's so ridiculous.
How like, like, sometimes guys and gals
and everybody got to touch them grass.
Like, yes, get off X.
Get off LinkedIn, let's get out because,
well, we're feeding our own like weird kind of deal.
And it's like, you know, you're right about that.
You know, like, I, do you guys know the,
he doesn't really, well, to my knowledge,
he doesn't really write anymore.
But do you, do you guys remember the column week,
but why, by turn Tim Urban?
Did you ever follow him?
Anyway, amazing writer, right?
And but it had amazing blog called Weight But Why.
And he was famous for doing literal stick man sketches, right?
Like he, it was just, I think he did him initially
and then maybe somebody else,
but he's sort of famous for that.
He'd be like, this is what's going on.
And it would literally be stick figures, right?
But his writing was amazing.
Eventually wrote a book.
Anyway, when I think about how he used to explain things
in his blog, he would offer talk about, you know,
sometimes, you know, we, if you're in it,
you know, your, your hill can feel very steep.
But if you extrapolate out to, you know,
in my case, almost five decades at this point,
then that, that hill is a really small bump
on, in other words, upwards to the right trajectory, right?
That's, that's the point of it, right?
So I think about that when you say that, Carl,
because, you know, we, we look at this, we,
I mean, we have a show about it, right?
But we, we look at AI and we do focus on the, the microscope.
We're, we're ants crawling over rocks.
And so we, every rock feels like a, feels like a,
a bit of a work, right?
And it goes up and down, it up and down.
And it feels very bumpy and very choppy and stuff.
But sure, right, most of the people we're dealing with are,
you know, human size, right?
And so like, they're not feeling the bumps in the road.
They're, they're, they're, they're putting their feet
on grass and be like, yeah, okay, that's, that is going on
that heard.
Also, what are we going to do about our ROI this?
Margaret, what are the things that really mattered today?
And so I, I agree with you.
I don't, I do not expect this to come up unless, for some,
for every once in a while, I have a client.
And they just like to nerd out with me on AI.
And then we might get into this.
Because they'll be like, oh, did you hear about that, you know,
it's like, okay, you're, you're at the ant level with me.
You're not, you know, you're not looking at it
from the 10,000-foot view.
You're, you're down here in the weeds weeds with me, you know?
There was, there was a partner, he's a partner
in an accounting firm we work with.
And there's, he goes, this is cool.
Oh, this is all cool.
This is amazing.
But at the end of the day, we're accountants.
So we need to do our accounting.
So, yeah, I don't know how we're going to do all this
because we're accountants.
And I'm like, that encapsulates what a lot of people
are like, this is all fantastic.
What do I have a job to do?
I have a family take care of.
So why don't you just talk to him about this?
But like you said, okay, now that's great.
What do we do about ROI in this corner?
Because we have a lot of stuff to do right.
Like literally this, it's Q, it's Q1, I'm in March, you know?
And I need to know what's going to change the trajectory
of my total sales by the end of Q1,
which is the end of this month.
And that's a very real thing, especially for me,
I deal with a lot of sales people.
So, you know, that, you know, Q, Q, your quarters,
your Q1, your Q's new results, they, they, you know,
they mean a lot, not only for obviously people
who make compensation too, because it can drastically
change their particular payouts and stuff like that.
So, yeah, I agree with you.
A lot of it is like, I like to geek out on this stuff too,
but also, and every once in a while,
I do have a like CEO level or whatever.
It's usually not that level.
Usually if I'm talking to people at that level,
it's like, what do we do yesterday?
What are we doing today to improve it?
And what do we think is coming tomorrow
that might affect today?
And then the really good ones are usually following up with,
hey, just checking.
You don't think we're painting ourselves at a corner, do you?
Meaning like, are we going too hard into one AI tool?
Or, you know, like, that's where an executive
is going to sit, right?
They're, they're like, are we moving forward?
Are we not realizing that we're moving ourselves
into a corner using this tool, using that tool?
And I love those conversations,
because then you could really like, you know,
you can get into the fun of it, you know?
Yeah, and that's what I think like most of the interactions,
there are some CEOs or executives
who geek out on this and they're really deep,
but for the most part, say, hey, I got a business to run.
I've got teams to go over.
I've got QBRs.
I've got OKRs.
I've got old bunch of alphabets and acronyms
and I've got a board like breathing down my neck,
that like, and then for non-tech companies, right?
They're like, crap, I got, I've got floorboards.
I got to make like, I got this stuff.
Supply chains like, hey, we've got trucks to move.
Like, I've got oil to ship like the like cloud co-work
with that, OK, that's great.
Can you move my thing faster?
I was like, you know, right, those are the conversations
you kind of have to just make sure it humbles you.
And it's sort of like, yeah, like this isn't the be all
and end all, yes, we kind of make it a big thing.
But like, yeah, there's other stuff that's going on.
And I think like, I'm sure there's bigger stuff right now
in the world, especially, you know, what's happening.
That is, you kind of see it and you're like,
and one thing I wanted to point out, I know I saw Trump
propose that note.
I was like, how many times has your president's flip flop
on to think so?
I was like, in two weeks, you'd be like, you know what?
This is too much of work.
And I think we all know, I don't know,
they're working in federal in government.
It takes so long to predict like, that's why once you get
like a software or something in, you can't take it out.
And you like are printing me for your company
because it's like all the bureaucrats, all the systems.
Once you've integrated, it's like, you can't pull that thing out.
Six months, I was like, that's going to take a year, my brand
because how many departments have like latched onto it?
I was like, how many people?
I was like, what are you talking about?
I kind of like this.
We're going to use this for the four months.
What do you mean we're going to switch to open AI?
Like, there's just, you've kind of,
like enough, like six months.
That's really impressive if you can do it now.
And I just want to name something in that
because the reason that it takes that long
is because you're interested in maintaining function
and support for your constituents.
Doge changed that.
Doge happened very quickly.
It changed some things.
There are some things that it said it was changing that it didn't.
But that's only a constraint if you care about an end result
that we think government should care about.
But that's not necessarily what's happening to it.
Yes, I know.
Like, I still think it's easier to unemployed people.
Like from a action perspective,
then it is to extract something that people use every day
that has become quote-unquote integral
to their workflows and processes and flip it to something else.
Because every single day you're still using it,
it becomes, I don't know.
I like to use like, sorry, it's common rate.
Yeah, venom, venom, venom, a symbiote,
like it just integrates with you
and it's harder to peel off.
Yeah, you come.
And it's, every day that it moves away
from just a simple chat interface, too.
You need a great straight, like, look through.
And that's just there, right?
I mean, I think there's, we're not talking
about anthropic at the department levels, you know,
hate, hate, hate, applaud, get in, get in, get in, get in.
We're probably talking about APIs
and very sophisticated ways of you.
Being misting is not, you're rad.
Please don't, like, I just, the joke's all there
with Pete Hanks, guys.
Can you lay out a 10 person strategy on how to,
it's like, hmm, high Pete, how are you doing?
Like, give me, here's a list of 10 things.
It's like, oh, my God.
He wasn't available for the press conference this weekend,
but apparently he is giving a press conference today
and it may already be happening, so.
Well, we'll keep following it up on this.
I mean, we've talked about this for about, well, sort of,
this, this and other topics for like 30 minutes.
I do want to say one last thing on that
that's just Carl, to your point about, like,
it's hard to move things in government and vet you,
you have lived for a while, right?
Around government, big, big government where you are as well.
Yeah, Kate.
I will tell you my experience of how I know this to be true
is when I was over in Iraq, yes, doing contract firefighting,
but I was there as a contractor on a base, right?
So for the majority of the time I was there,
I was on the Victory Base Complex,
which was a massive base complex
that had smaller bases, striker,
and so it ingrained with it, right?
Right in and around the Baghdad airport.
So the fact that Baghdad airport was on the base.
Let's put that one.
Anyway, here's my, here's my answer tonight.
Here's how I know things are so overly complicated
when it comes to ingrained in the government
to get a new ream of paper for your printer.
Like if you just wanted a new box of a printer paper,
was an act of Congress.
It was insane.
The amount of steps you had to go through
and you think, man, if they would just give this much detail
to, I don't know, education or like literally anything else,
we ever designed this insane system for how I get,
you know, requisited it and I'm allowed to,
but I had it, but I had a show proof that we were out of paper
and that like we weren't just ordering this stuff
because that was a problem, whatever.
Anyway, I could tell you a story upon story
about story about office chairs, forget about it.
Your office shirt could fall on the floor with three legs.
And like maybe in three months
we'll get you something to sit on
and the meantime goes to it on this stool
because you're not getting a new office chair.
That's, we would go dumpster diving for office chairs
and try to replace them ourselves.
It was faster, you know, this anyway, that was an eye open.
I think that for anybody who has never worked for government,
of course, I was domestically here
and I worked for government locally
and that's very similar too.
But home moly, when it comes to the government level,
at least in the US, that is for everything.
And that form needs to be signed by however many people.
Oh, it's good luck if you can get the call to sign it
in whatever timeline you, it was a half a day.
It was a half a day of walking around the base
with a clipboard.
That's not even a joke.
It was a half a day.
If you were lucky and they weren't out to lunch
or on vacation, right?
They weren't off base or whatever of a kitchen.
Okay, just wanted to bring it up
because like I have like the brazenest memories
of trying to get stuff done
or moved forward on that base over my 15 months.
Okay, I do really want to bring this up.
This is kind of a new story
but I think all I want for all three of us to talk about it
because I'm curious what y'all think.
Okay, Nate Jones.
All three of us have watched, have talked about,
we all know Nate Jones.
If you don't know Nate Jones,
he does put out actually really good content on probably X
but I see him on TikTok.
That's where I get along with him.
He has a really RSC stuff
and he also has a paid for a substack.
So I can use emails but all the times his emails
are just like set ups to go to his paid stuff
which is fine, he's got a funnel.
This video, in the fact you can use it,
you can see if you're looking in the first comment right here
like wanna read Nate using that AI for content now.
And then somebody else says this is an AI video.
Somebody else says yes, uncanny Nate, AI Nate.
And I do find this interesting because before I read the comments,
I thought I think this is an AI video
and I find that odd because that's not what Nate does.
It's on his deal so that's pretty strange.
And it is him, by the way, this is his account.
This is Nate Jones.
So this is not like somebody put this up
and was pretending to be in with his voice or something.
So let me just play a little bit of it.
I don't even think you, I mean, you can look at it,
you can listen to it, listening probably wouldn't do it.
But if you'll guys, because you guys both know him,
this doesn't have his natural pauses.
And I think that's a, they do jump clips
which I feel like kind of messes with you a little bit.
But anyway, let me just play just a little bit of this.
Educational philosophy decide at best
for a 20th century industrial economy.
That economy is not gonna exist
when these kids get to adulthood.
This is like the calculator moment,
except it's for everything.
Back in the 1970s, when electronic calculators
became more affordable, the education establishment panicked.
Calculators in classrooms were considered cheating,
full stop.
They would destroy children.
Okay.
All right.
If you look at his eyes, he doesn't blink.
And that's when people were calling out of like,
it's in the eyes.
There's something uncanny about this.
It's, he's farther away from the camera
than he typically is.
We can just look at it.
It's a free, a work form.
This is a normal, yes.
This is not an AI.
I gotta be honest with you.
If you're looking at reading books, I can read it.
Just different, right?
He's looking down through his glasses.
I know, what do you guys think?
Would you say that first video is AI?
I, what I, my response to the first video
is that he's, it's, I don't know that it's AI,
but he's using, like, look at the camera thing, right?
Because like eight, where your eyes don't move away from it.
Yeah.
When Nate does videos, he's got something like right up there
that he sort of just slightly looks at
and then comes back to give the next point.
But he may just be trying some different things.
I have no, that's a more extreme jump cut.
But he does, like his videos,
sometimes I have a hard time listening to his videos
on, on like 2X or 1.5X
because there's no breath sounds.
And because the breath sounds are edited out, right?
And so, and then when that goes faster,
it really sounds like somebody's doing this,
never talking, never breathing.
And I'm just like, okay, I know.
That actually gives me anxiety, by the way.
But like, I don't, that actually,
I could feel my, like, my heart, like my, it in my chest.
So I can't really, it's some people like those,
because they're painfully slow and so 2X works.
But I've actually done that last,
it's funny you mentioned that because so many people
edit out all the breaths, all the sounds.
I don't like that, we don't do that on this show.
I don't want to do that on this show.
I think you need to hear the spaces and the breaths.
And you need to hear us all go silent
and then it'd be like three beats and then Carl go,
yeah, I mean, I think it's this.
I don't want to take about that three seconds.
That's, that's all trying to think about what the answer is.
You, that's important too.
Like that the silence is important to the conversation.
We also do a live show.
He's doing curated content.
I get it.
I will tell you, there were so many people
in that comment section saying,
this is AI, right?
And also like, that's weird that Nate didn't call this out
because, I mean, I'm a follower of his.
I watch his videos.
I think he's got really good content.
I don't even know that I would care necessarily
that he was like, hey, trying something new with some AI
if that was the case.
But I think personally, I'm kind of in the in the camp of,
I mean, yeah, dude, do it, but, but say you did.
Yeah, yeah, you know, it kind of bothers me, honestly,
a little bit because like I do consider him
a nice source.
Now, I guess I don't know for sure.
And I do think it's weird that he is,
he's since put up another video,
which is probably the one I just clicked to.
Yeah, that was the next one I was just showing.
And there's, there's no mention of it.
So I don't know, but I do find this surprising from him.
I've never seen him do this before.
He's always a straight to camera.
Look at this, look at that leaning in.
He's usually got the beat on, you know, like he has a,
he has a certain look and, you know, feel of his videos.
And that does feel uncanny.
I don't know, what do you think, Carl?
The editing was it, that cut from one,
there was one cut there that was a little wonky.
And I was like, if you're an editor,
whether you edit or somebody else edit,
you normally wouldn't make that mistake
because you're just cutting, like, I'm gonna cut this section
because I screwed up or whatever.
There's that one little section, it's cut, you know, like,
that doesn't look no editor,
whether it's junior or brand you ever make that mistake.
That would not be a mistake.
You just, it's, you can't, you don't do that.
It's very strange that would be a weird kind of,
yeah, it's kind of interesting on that screen.
Just me y'all can see too.
The one up here, the second from the top, from the left,
that's the one, look how much further back he is
than every other image.
This face is right there, right?
It's weird.
And the other thing that you can see,
and I have not clocked this on things
that were based on photos,
but you can see that when he gets excited about stuff,
the, his facial flush changes, right?
Like, there are noticeable differences based on how he's
feeling about what he's talking about.
And that may be one of the indicators
that we start to see in AI avatars
because, I mean, they're just, they're trying to make sure
that your lips look like they're saying the sounds
that people are hearing.
That's like the level of what they're trying to do.
And like trying to match performance
in the video performance when you're gonna face swap
something, but nobody is like yet actually getting to be,
oh, okay, so we finally got what this emotion should be
and what it sounded like.
Now we need to match that to what the eye shapes are doing
with the facial flushing.
We're not there with avatars yet.
Yeah, you look, and by the way, in the comments,
I'm just, I was just curious.
I think Jeff saying, I mean, there's definitely maybe
an AI filter.
So on his eyes, so that, so that in a period,
he looks at the camera and he can read the script.
So Jeff is saying, maybe it's just,
it's just like the eye tracking thing.
Gareth is saying, it's always the teeth.
I don't know if it's AI all the way.
Let's see, Jeff is saying, yeah,
I just think it's a filter.
Yeah, so listen, here's the thing.
This stuff exists.
I'm not against people using AI to help them create
more content as long, I mean, I literally do that
with a conundrum episode every Saturday.
But it's literally the thing I say is this is gonna be
two AI go, so I'm not trying to be like,
sit in my high horse or anything like that.
I'm just saying that like, it is what it is.
I don't think it diminishes that conundrum episode.
I think it's still a good conversation.
I still enjoy them, but I'm just telling you what it is.
And when you have somebody like this who is,
is prolific as posting as neat.
Maybe I'm wrong, but it wasn't just me.
I thought it was AI, looked in the comments
and there was a lot more people.
So he either used the filter, he changed something up,
just to try something different.
But it's, and it, by the way,
this isn't really even about Nate Jones to be honest with you.
We're giving him a lot of, a lot of repress here on this show.
I think it's just a, it's the larger conversation.
That's why I wanted to bring it up today
because it's just a larger conversation of like,
you know, what if this was Rachel Woods?
What if this was, you know, other people,
Hallie Miller, other people that we follow.
I don't ever think it would, but
Ethan Malik, you know, it's not his thing,
but I don't know, what do you guys think about this?
Like if somebody does this
and let's just say this was AI, hypothetically,
I'm not saying I know, would this bother you?
So I'm wondering what are the scenarios
in which this would make sense, right?
Nate had a, Nate's a solo printer, his face is the,
is his brand, right?
If Nate had a situation for any kinds of reasons, right?
An accident, a hospitalization, an emergency in whatever,
what would you then turn to to try to keep your schedule?
Like those kinds of things make sense to me
and it would make sense to me
that you wouldn't necessarily put out,
you wouldn't have ready, hey, I've thought of all contingencies,
just like I have with AI, which I mean, he's like absolutely,
the guy is thinking about everything in terms of AI,
but it's different to turn around
and put that lens on yourself
and think about why that situation would happen.
He also does a crap ton of research,
and that would make sense that this was a research thing
and like, hey, 78% of the audience,
that's what I think, and you might be right on that,
that would not, I had that same thought was,
what if he comes back in three days and said,
yeah, guys, I did this on purpose, whatever,
and I mean, honestly, he could have that out
whether he did or didn't, he could like probably,
because he could, you're right, he could pull that off that,
like that would be a thing Nate would do.
And what I said in the comments too is like,
isn't the bigger conversation that we have,
we have us, I'll call all of us intelligent,
I'll throw myself in there,
we have intelligent people in the audience,
and nobody is saying for sure
they definitively know one way or the other.
And isn't that really the bigger conversation is like,
here's somebody who has a plethora of,
stock of example footage of live footage.
We're looking at 300 videos to one.
I've probably watched every video of his on TikTok
over the last several months.
I don't usually skip them, those in Harry Mack videos.
That's another conversation,
but I don't skip those videos.
I want to know what they're talking about,
but yeah, I don't know.
Anyway, it was just a conversation.
I just curious with you guys more,
like not really like news news,
but it was definitely something that hit over the weekend
for me and I was like,
oh, I want to talk about this on the show.
Yeah, and just to say,
so there are lots of people,
Garrett who wears with him or has engaged with him
for a lot longer, he's saying yes.
From what I know, he is very authentic.
That's my experience with him too.
JDE Salt isn't as pulled from a much longer version,
like almost 30 minutes, right?
Yes, the snippets are intentionally created
so that they can be used in the short world,
but he's normal videos are much longer.
Yeah.
So, okay.
We love you Nate.
Sorry if we stirred some shit.
Well, listen, I mean,
I'm not the only one saying this.
I wasn't the only one bringing this up.
And you know,
I think it's a great conversation to have
regardless of whether it's sure or not,
because I would love to hear Nate's thoughts
on why people thought it was AI if it's not.
You know, like they ain't gonna let's talk about this.
Let's have a bigger conversation.
So I'm looking forward to something.
Right, yeah.
Yeah.
So we're looking forward to the rest of the story.
Gary's saying he thinks probably Nate used an AI editor,
which again, like this could easily be
from the script editor, right?
The underlord, overlord, whatever the script has.
Like make sure that my eyeballs don't ever leave the camera
and then you're just like this.
I mean, it's like you don't want that to be your video.
That's not a good look.
Yeah.
I don't mean that from Nate.
I mean that from like the script.
And the people who sort of apply it indiscriminately.
I mean, we, I used it somewhere like,
I mean, this guy, when Adobe came out with a memory,
Adobe came out with that like two years ago,
the eye tracking thing.
Yeah.
But the problem with it is you can't,
you really got to, like it's fine if you're talking head
and you kind of stay still or whatever,
but no, yeah, most people called it a man,
you know, it's more distracting than not.
Like anything else, if you have a 15 second clip
that you're doing, you, the shorter the click,
the more you can get away with AI
because there's just not enough there to say,
oh, that's weird, and that doesn't sound like Brian
or doesn't seem like his mannerisms.
I move around a lot.
Have y'all ever noticed it?
If I ever, if it's ever me as an AI avatar,
it's gonna have to get the lean.
Like I'm always leaning on one.
I'm always over here, over here,
which I noticed early on with Jimmy,
would I be like, can you, can you stay
in the middle of your frame?
You know, like on the side because he was doing the editing
and stuff like that.
I was like, oh, I lean.
I like my grandpa was like that.
He would always lean on the, on the side.
So my shoulders are always crooked.
I'm not squared up to the camera.
So if there ever is a AI avatar
and you guys are watching the show,
you're like, I don't know, you'd be like, listen,
that I, I avatar for Brian,
that I have a damn lean on it.
Otherwise, it's not true, it's really him.
Yeah.
So I just, I just want to pull out one comment
because I think it's, I think it's a good thing
for us to respond to, it's a new commenter,
DiskoLot saying, this is an AI witch hunt.
I don't understand why or for what purpose.
So what if he did?
I agree.
So what if he did?
But I mean, it's more the conversation I think about,
like, it's not a witch hunt.
I'm not trying to take down me.
I don't, I just, as a fan of me,
I wish Nate would have said that he did it.
Maybe he will, that's it.
Right.
I'm not a witch hunt.
I'm not trying to take down me.
I like his content.
I will continue to follow him personally.
I think it just, like what I was thinking about was,
so Matt Wolfe recently said, hey,
my marketing agency has done this, like we're talking,
he was, he, I think it was a comment he was talking about,
like, he's hired a marketing agency for himself to,
you know, because he has such a good audience.
How do I monetize?
I'm sure.
How do I monetize this audience and so on?
It would be kind of, it would be very smart
for Nate to do the exact same thing
because his audience is growing.
How do I, you know, monetize my audience, like,
and, and, and, and do that?
So whether, I think it starts becoming an issue
if you're, and I don't think this is Nate,
but like if you're brand or how you portrayed yourself
is offent, is like, is center round offenticity, right?
Personally, do it because if you don't,
then, like, then, then it, it separates from you
from what you were, you know, you put yourself like this show.
Literally, if we did that, I think people would have a right
to, you know, have an opinion about whether we, you know,
I mean, if we did it up front, that's one thing,
but at any point, if we're, if we're interjecting AI
into these conversations beyond the,
let me check really quick and come back with the answer,
that kind of stuff like a Google search.
If we start using AI more for, like,
heavier editing live or something comes out in the future,
and I don't know what that might be or whatever,
but there's an insert of, you know, that's not all real,
I mean, we're a live morning show five days a week.
People should have an opinion about that
if we decided to do that.
So I don't know, that's kind of what you,
I don't know, it's felt like what you're kind of saying, Carl,
like, we're, we're chill built on authenticity.
We say dumb shit and we say good stuff.
Sometimes we say really good, really, you know,
million dollar ideas and there's a lot in between.
That's the show, you know.
Yeah, so yeah, much longer conversation.
I suspect that there will be follow ups to this.
I think we expect that Nate will talk a little bit
if he hasn't already.
He has a really great chat section
in his substack and he may have already addressed it there.
I've not been able to go in Jeff, assume I'm using AI
if you see me on camera.
Yeah, and I think I get the witch hunt,
the thing and we're really not trying to do that.
We're really just trying to surface the story.
And I'm not trying to witch hunt anybody.
I'm not against people using AI and clear, I mean, I'm not, you know,
again, if it's quality content, it's quality content.
I either got something out of it or I didn't.
It's my choice.
No, and it's relevant.
Pika just released Pika selves, Pika AI selves.
I do not know whether these are avatar likenesses
or whether they're agentic something that you can put in
in a Pika self, but that's like something that's bubbling around,
around X and agent keeps updating their stuff every, you know,
a couple of weeks, a couple of months.
So there are more opportunities for this kind of thing to happen.
And I think what we're responding to is the thing that you may not be thinking
about that gets noticed, right?
Jump cuts are not a thing because AI did them.
Jump cuts are an editor choice, but your human editor may be paying
a little bit more attention to what Carl was said.
That's not something a human editor chose to put in that clip, right?
I probably have to jump, but like, do you guys want to keep going?
Do you all want to keep going?
And I can bounce or we can start wrapping things up for the day and just pick
up tomorrow with more of the news and things like that.
I think it's been a good conversation though on, you know,
less, less topics, but more in depth on, you know, on today.
What's interesting is, yeah, I, I don't know, like, I haven't seen outside
of the stuff we talked about the rap.
I haven't seen any like major topics that have come up since then,
great, or releases or anything.
Oh, it's just like Friday.
Yeah, we're since outside of what we already talked about.
Yeah, the two, the two I saw were that X,
Grock was going to like allow people to do their own voices,
like they're going to be able to do voice synthesis.
I saw that one.
And then Microsoft leaks copilot canvas, which I was like, I mean,
it's, I don't know, that's not record breaking news.
It's just, I guess they have like sort of a canvas workspace, freeform workspace,
where I do like that idea, by the way, this idea of a freeform workspace,
where it's sort of like copilot's there as it's instant images.
I don't really do a lot of like whiteboarding or, you know, like vision mapping,
but I feel like for people who do, that's got to be super cool.
If you just have a blank space and you can vision board with AI,
a picture that's a really cool experience.
So if that's kind of what they were talking about, I think, yeah,
that could say a lot of people using that to just like get a group of people
to gather either virtually or in a real room and throw it up on the board.
Would you have AI helping you with the images and all the stuff?
That's cool.
Yeah, absolutely.
I feel like that has been something with some of the other boards.
I have a bunch more things, but I'm okay going.
One of the things that I want to surface before we do, Carl, is you referenced it
in what you were saying.
There is some leakage about a chat GPT 5.4.
Are we skipping numbers?
Like the thing is though, like we, I think we've seen so many leaks.
And it's so easy to to punch whatever the leak is.
So it's sort of like, yeah, there's a bunch of like numbers and stuff going around,
but it's like, unless they actually release it, we don't, we don't, we don't, you don't know.
Well, there was the 5.3 code X wasn't there.
So yeah, I'm going to code X.
And there is the 5.3 chat GPT we assume.
I mean, like they've talked about it, but there was a buzz this week.
And in 5.3.
Yeah. Well, listen, going back to our original story, I mean, what's a great way to get people
talking good about you release the next best puddle that's a close point back around me.
Go, that should be the reason to look what it can do.
You know, so they didn't do that over the weekend.
But if, but if AI ever wanted to dangle a carrot over here and be like, Hey, look over here.
Instead, they, they have the ability to probably do that.
So I guess you have to find out what happens this week.
So all right.
Well, let's do that.
Let's go ahead and wrap it up for today.
I think the three of us are back tomorrow and for parts of this week.
And he's still out for a little bit.
But we should have a Jimmy and Ann maybe in at other parts of the week.
So keep coming back, keep hanging out with us.
Thank you, by the way, to everybody in the comments.
I mean, we had a lot of people today.
Our live numbers went really up as well.
I really appreciate by the way, there's a direct correlation by how much is going on in the conversation
in the comments and then how many people are joining us live.
So you guys by having conversations, by interacting with the show on YouTube,
are directly sharing and letting other people know to come see the show as well.
And it may be their first day.
We certainly had a few people today who, you know, had had more recent or I hadn't seen them before on the show.
So welcome to anybody new on here.
We appreciate you all.
This is the time to let you know we do this show five days a week.
Monday through Friday, 10 a.m. Eastern every single day.
We have not missed the Monday through Friday since we started.
And this is episode 671.
So later this year, we'll hit our three year mark and we're rocking the roll in.
So with that, we'll say goodbye, have a great day.
And we will see you all tomorrow on Tuesday.
Till then, bye.

