00:00:00
All right, everybody. Welcome back to the number one podcast in the world. It's your favorite podcast. It's your
00:00:06
podcast's favorite podcast. It's the All-In podcast. Episode 279. With me, Fredberg, Sax, Chimath, you know the the
00:00:14
squad. You know the squad. We're here. It's the summer and we're ready to rock and roll. We got a power docket. We got
00:00:20
a rocket docket. Palunteer and Nvidia have announced a sovereign AI partnership. Where have we heard that
00:00:26
term before? Palanteer is going to use Nvidia's Neotron. Neotron like the Pixar film open models to build a custom
00:00:36
frontier quality model to serve the US government. Palanteer is calling this new platform sovereign AI operating
00:00:44
system. The US government agencies will own the hardware, the data, and the model waits. Palanteer also shared a
00:00:50
viral tweet manifesto laying out the concept, "Data retention is your treasure. Transfer it at your own peril.
00:00:56
Transferring that data hands over access to your pre-existing winning plays and yields the means of production for new
00:01:03
ones. CEO Alex Karp went on CNBC to announce the partnership in a classic Karp Robin Williams style monologue. Here's a
00:01:14
clip from his 20inut interview where he where he basically went after the frontier models like Anthropic. Play the
00:01:23
clip. Our clients are just to say they're unhappy with the Frontier Labs is to say I'm welcome at the Berkeley
00:01:30
faculty. It's like there's just a level of discomfort and loss of trust. Sam and
00:01:36
and and Daario, there's nothing more fun than debating Daario in private. So this
00:01:40
is I'm not throwing shade at them, but something has gone completely wrong. And the basic view among enterprises in this
00:01:48
country is I'm going to chill lax uh and waste my time with tokens. I'm going to
00:01:53
get no value and they're going to get my IP. When the do department of war goes to you and says I need this application,
00:02:00
do they get to control the weights to do it or do you get to control the weights?
00:02:04
Are we really going to outsource the battlefield of this country to the consensus view in Silicon Valley? That
00:02:11
is effing insane. >> All right. And so some folks uh refer to this as a televised nervous breakdown.
00:02:19
We here at Allink call that Alex Garp on a Tuesday. And we talked about this Chimath uh Sax
00:02:26
Freedber. We talked about this a whole bunch back in February. I coined the term intelligence
00:02:33
sovereignty here. Here's your victory fact. Do I want to give all of the secrets in our organization, every piece
00:02:40
of intellectual property to Sam Alman who's got to make a billion dollars a year to keep up with his spend, right?
00:02:46
He's going to build every application. I've been talking about AI sovereignty here for a bit just in terms of how much
00:02:52
more coste effective it is and how you're not training other people's AIs with your knowledge and your insights.
00:02:58
This is why it's super important that open-source open- source agents and local hardware be able to run these
00:03:04
models and that consumers and companies learn how to roll their own language models. The intelligent sovereignty is
00:03:11
different than privacy. Privacy is oh, you can't see my photos. You can't peek into my notes app and what I wrote there
00:03:17
in my journal. Intelligent sovereignty is you can't tell me what to think. You can't use your AI to analyze my photos,
00:03:24
to analyze my emails, to analyze my messages, and tell me how to interpret the world. That's actually going to be
00:03:29
the next key piece here. All right. And uh Sax, you are in your long post era. Another long post this week from you on
00:03:39
this very topic. And obviously as uh AISAR there for the first half of the Trump administration, you uh been very
00:03:47
involved and very close to this. I would love to hear inside I'd love to hear your take on this and your long posts
00:03:53
over 300 words. You can follow x.com/davidsax, but also like the palace intrigue here
00:03:59
and what this means in terms of the relationship with the government, which we're going to get into in our
00:04:05
second story. Well, look, Jay Cal, I got to give you some credit there. The first part of
00:04:10
your take was spot on. After that, it was kind of diminishing returns. I'm not sure why we had to listen to the next 30
00:04:16
seconds of it, but anyway, uh, it started off really strong. But let's go back to the supposed crash out by Karp
00:04:23
on CNBC. It was nothing of the sort. It was all these legacy media types making that claim. And that's the first clue
00:04:30
that he's actually saying something insightful and maybe kind of brilliant. And I think the thing that he said that
00:04:36
I hadn't really thought about in quite those terms is he started talking about AI safety in the enterprise and what
00:04:42
that really looks like. And what he said is that what technical customers want is
00:04:47
control over their compute, their models, their data stack and their alpha, meaning their proprietary
00:04:52
knowledge. They want to know they own the means of production, he said, and it's not being transferred to someone
00:04:58
else. And what he's referring to there is that these enterprises are at risk of transferring their knowledge, their
00:05:05
knowhow, their trade secrets, their customer data to these model providers who might eventually decide to compete
00:05:11
with them. Like you said, JCAL and you can see that enterprises are waking up to this threat and they're not happy
00:05:18
about it. And I think KARP is exactly right about that. Now, I think this is a really interesting
00:05:26
take on AI safety because what safety means for an enterprise is again that they get to control their own data,
00:05:35
their model weights, their compute. So, a frontier lab can't hoover up their proprietary knowledge, their alpha and
00:05:42
turn it into their next product. And if you don't think that can happen, just look at what happened to Figma. So
00:05:49
according to the information, Anthropic quote unquote blindsided its then business partner with the launch of
00:05:57
Clawude Design. So this was a new vertical app that Anthropic launched to compete in the design category. And
00:06:05
Figma's founder said that Anthropic had not been completely honest with them. Enthropic chief product officer had
00:06:11
actually even served on Figma's board and didn't resign until 3 days before the launch of Claude Design. So
00:06:17
obviously Figma again felt blindsided by this and you can see the resulting impact on their stock price. Figma's
00:06:24
stock has fallen something like 50% this year while Anthropic's valuation has surged. This is not an isolated example.
00:06:31
Anthropic has also launched Cloud Science, Claude Security, Claude Legal, Claude Financial, and of course Claude
00:06:38
Code. And every single one of these vertical apps expanded into categories that was previously served by companies
00:06:45
building on top of Anthropic's own models. And really, if you want to go back to when Anthropic's revenue
00:06:53
explosion began, it was with the launch of Claude Code. And how did they know to
00:06:57
launch that product? Because they saw that cursor was doing extremely well. Curser was one of their biggest
00:07:04
customers. They created the coding assistant first. they created that category and anthropic said oh like why
00:07:11
don't we vertically integrate so in other words they're watching where the value is being created on top of their
00:07:16
models then they're moving in directly and this is a formula that I think is very Microsoft like you could say it's
00:07:22
very Google like they want to dominate the model layer you could call that the operating system and then use that
00:07:28
position that monopolistic position to capture the most lucrative verticals so if you want to think about like the
00:07:35
Microsoft example they had the Windows monopoly and then systematically they went and dominated every lucrative
00:07:42
category of business software. It started with spreadsheets and word processing and then eventually it went
00:07:46
to the browser so forth and so on. If you want to look at Google, they basically had a monopoly or dominant
00:07:52
position in search. And if you go back to the early days of Google, the search results picked you off site. And in
00:08:00
fact, they really pride themselves on how quickly they could send you off site. But gradually over time, they use
00:08:05
that traffic to tell themselves where to build properties. And today, fewer than
00:08:10
half of searches kick you off site. You stay on Google properties. And I think something similar is happening with
00:08:16
Anthropic here. the pattern is clear, they are going to use their dominant position in the model to then grab more
00:08:23
and more territory in any interesting and lucrative vertical. So again, back to Alex Karp's point, if you're an
00:08:29
enterprise customer or a developer, why in the world would you ever want to share any proprietary data with them?
00:08:35
You are mortgaging your future. You are sealing your fate. You are going to lead
00:08:39
to disaster for your company. Just one last point then I'll turn over JCL is that Daario at the same time that they
00:08:45
pursue this business strategy has been arguing that open- source models are dangerous and need to be restricted.
00:08:51
Well, dangerous to whom? Not to enterprises that want to retain control over their data.
00:08:57
>> It's dangerous to his business model because his business model requires >> that customers don't have a lot of
00:09:03
choice at the model layer. And what Karp is pointing out here is that if you want
00:09:08
to have true AI safety as an enterprise, you have to retain the ability to choose
00:09:14
at the model layer who gets to see and use your alpha. >> Yeah, this is well said. I think you you
00:09:22
picked that carcass to the bone a bit, but Chimath, you're actually doing specific examples of this uh at 8090.
00:09:29
You've been testing uh some of the open source models. I saw you share that on Twitter X. So maybe you could give us a
00:09:36
little feedback on what you've learned as uh the CEO of 8090 and you know your firsthand experience now with using open
00:09:43
source for the first time in the last couple weeks for this specific use case in enterprises. When you start a company
00:09:48
I mean you guys all know this you're not starting it for the moment that exists today. you're almost sort of
00:09:55
trying to forecast if such and such a set of things happen then here is the scene that that gets created because it
00:10:02
takes time to build something and it takes time to get enough reps to know what you're doing in your go to market.
00:10:08
This for me was the moment that I thought would arrive which is the point where everybody wakes up and realizes
00:10:15
wait hold on a second two things are true. The first is that my business is complicated. I want AI to be able to
00:10:23
accelerate it, but I want to be able to protect myself in doing so. And then the
00:10:26
second is I want the flexibility where there's an independent third party control plane that I use to get all
00:10:32
these benefits so that I don't leak and seed my advantages away. And I think Alex is an incredible,
00:10:41
smart, brilliant guy and he completely nailed it. And I think he called out on its face the huge risk of this. So let
00:10:49
me just give you this narrative in three tweets. The first one is I read this really interesting study from BCG and
00:10:57
what they looked at was the return on capital employed or ROCE of various businesses. And this is what's
00:11:05
incredible. The cost of capital has now with long-term rates moved back to what its long run average
00:11:13
is which is around 8 to 11%. What that means is like that is the actual cost that you would borrow money at
00:11:21
effectively. The problem is that half of large US companies now cannot deliver returns that exceed that. That is a
00:11:29
really big problem. And then second there's a further problem which is that persistently low returns. So in the you
00:11:35
know 1 2 3 4 5% is about one in seven companies all around the world. Okay. So why is this important to note? It means
00:11:43
that being in business is complicated. It's hard. Not everything works all the time. There's a bunch of underperforming
00:11:48
businesses. There's a bunch of underperforming segments. So in that lens, when you, you know, think about
00:11:54
what Sax said, which is you have this company that comes to you and says, I have a magic box and all you have to do
00:12:03
is tell me everything you're doing, and this magic box will make everything better. But then all of a sudden, from
00:12:09
the shadows, the magic box says, you know what, I've decided to compete with you. That is a huge risk. And now that
00:12:15
you've seen enough examples of it, I think you have to figure out a different way to do it. So then you go to the next
00:12:20
tweet that I saw, which I thought was interesting. And this is a woman who's an ex meta PM. And what she essentially
00:12:27
says is like, hey, hold on a second. There is this assumption that you can't use an open- source model because it's
00:12:33
all Chinese. And what she says was, well, even if it's 100x cheaper, and their response
00:12:39
is, no, because we care about safety and security. And her perspective is, don't
00:12:43
you understand that you can actually host open- source models with your own GPUs in US data centers that doesn't
00:12:49
share any data back to anybody? And instead, what you're accidentally or purposefully doing is giving away all
00:12:55
your data to a couple of frontier labs rather than owning it privately yourself.
00:13:01
And 100x, it turns out, is a really big number to pay to do all of that by accident.
00:13:09
And that's what Alex Karp is saying. He's like, why would anybody do this when there are alternatives? And so the
00:13:15
third post that I'll talk about is something that we did. So we took our software factory which is an agnostic
00:13:21
third party control plane and we just wanted to see and we ran it on a very typical enterprise task which is you
00:13:30
have an old piece of code you want to migrate it and you want to maintain it in a new framework so that it's easier
00:13:36
and more flexible. pretty straightforward task. And so we ran it and we ran an experiment where we did
00:13:44
Claude by itself, then we did us plus Claude, and then we ran it on the best frontier open-source model and then us
00:13:53
plus that model. And the data is crazy. So when you use our harness with Claude,
00:14:00
it was simultaneously 1.4x 4x cheaper and 1.5x faster than just using anthropic opus 48 alone. But if you wrap
00:14:09
the open source model with our software factory, it was 16.4x cheaper. Now it was three times slower. But you know,
00:14:18
you're talking about a couple of extra hours to save 16.4x >> on that one. The slowness Chimath, is
00:14:24
that slowness because of the hardware being served up by Claude or is it >> This is all using open router. know this
00:14:30
was all using open rudder on a very traditional hardware stack. So >> look I think the reality is could that
00:14:36
be optimized even further? Absolutely. But my my point is if you take Sax's points and then if you take Alex Karp's
00:14:42
point and just this actual data there is a very legitimate question which is if you are a reasonable company
00:14:50
why are you not finding an independent way to access this intelligence in a way that doesn't leak your edge away to do
00:14:56
so at this point now is kind of becoming derelict and irresponsible. Back then you could be experimenting because you
00:15:05
didn't know any better. But now when you know all of these data points to continue to make the same decision I
00:15:10
think is really insanely dumb. All right Dave Friedberg you are also a CEO of the
00:15:15
surging oh and you have a lot of proprietary data. Let me ask you point blank. Do you trust your data to the
00:15:24
frontier model companies or are you doing what there seems to be consensus here protecting your data protecting the
00:15:31
crown jewels so to speak and using open source are you experimenting with it and
00:15:37
just yes or no do you trust the frontier models with Ohio's data so I'll tell you
00:15:42
there's been an effort by anthropic to go around and sign up life sciences companies to contributing to a new life
00:15:50
sciences focused model that effort has been they're approaching these large companies with large proprietary data
00:15:58
sets and saying hey if you share your data we will give you early access some sort of proprietary value sign this NDA
00:16:04
and you can participate with us and I think nearly everyone I've spoken with has woken up to the fact that they are
00:16:11
basically trying to commoditize everyone's business because fundamentally if all of the tens of
00:16:16
billions of dollars you as a life sciences company have invested in experiments and product development
00:16:23
and you've generated all of this proprietary data along the way. That data is a true asset of your
00:16:29
organization. It's an asset that you've spent billions of dollars developing. And by handing it over to a model
00:16:35
company to then combine with other people's data, you are effectively commoditizing the asset that you have,
00:16:42
the one kind of core differentiation that you have. And so everyone is largely saying no. The way I see this
00:16:48
evolving is very much in line with what Alex Karp suggested on CNBC. If you go back a couple of years, I think we all
00:16:56
assumed there was going to be this large hub, large spoke model for AI model development and deployment. Meaning
00:17:03
there would be these very large clusters. These large clusters would be ultimately capital advantaged. So those
00:17:10
who had the most capital, which is why everyone's raised tens and hundreds of billions of dollars, would be able to
00:17:15
train models. And then there would be these large spokes, these large clusters for deploying those models with
00:17:21
inference. So everyone's using the NeoClouds and the hyperscalers and whatnot to run models. And then maybe
00:17:26
they've got their own proprietary data layer that sits in front of that. But I think what everyone's realizing is
00:17:31
they're better off developing their own weights and their own models using either an open-source basis or there
00:17:40
might be some intermediary business model that evolves. Meaning there will end up being several large hubs that do
00:17:46
all of the core foundational model development. Then smaller hubs meaning like clusters for training that
00:17:54
enterprises will use to train and develop their own proprietary advantaged models using their own data and then
00:18:02
there will be these much more distributed spokes because I think everyone's also realizing the value of
00:18:07
onrem by putting a set of servers and building a cluster in your own data center or even in your own enterprise
00:18:14
you know IT closet you can run a lot of the workflows that you're using AI for for your enterprise locally. And so I
00:18:22
think the model is shifting where we're going from large hubs, large spokes to large hubs, medium hubs, and then a
00:18:29
distributed spoke model where there will still be, you know, NeoClouds and hyperscalers that are being used for
00:18:34
inference, but people will also have their own inference instances that they're going to run for their own
00:18:40
enterprise setting. And everyone I think is walking this path and they're going to walk this path over the next couple
00:18:44
of months because they're realizing quite quickly that in order to compete in a world of commoditizing knowledge
00:18:50
and commoditizing capabilities, you have to leverage the core differentiating assets that you
00:18:57
have, which means you have to build your own models and you will likely end up having to run your own inference with
00:19:01
your own proprietary models. >> Yeah. And to just give a a little bit of texture to how that's going to look, if
00:19:08
you follow the Microsoft example, which we've talked about here before, Lotus 123, Word Perfect were their partners.
00:19:14
They were replaced with Excel. They were replaced obviously with Microsoft Word.
00:19:19
You don't even know those other two. Word Perfect, Wordar, Lotus 123, Vizzyalc. That's exactly what they have
00:19:25
to do. And they have no choice but to do that now because they have a trillion dollar market cap. They must win the
00:19:31
application layer. And Sam Alman went to Y Combinator and he said, "We'll give you $2 million worth of free tokens."
00:19:38
And I came out and I said, "Listen, there's nothing personal against Sam. You know, Sam's a very aggressive
00:19:45
dealmaker." And he wants to get access to those startups cuz he knows having run Y Combinator that if he can get
00:19:51
their innovations, those founders latest thoughts about what's around the corner,
00:19:55
he can incorporate them into the platform. There is no free pizza. There's no free beer. When somebody like
00:20:00
Sam Walman comes to you and says, "Here's some free tokens. Your alarm should go up." Zuckerberg did the same
00:20:06
thing. He said, "Hey, I'm going to give people a bunch of access. Going to give them money. Come to the Facebook
00:20:10
platform." Nobody who went to bed with Microsoft in the 80s, Facebook in the 2000s, or Sam Alman now in in the 2020s
00:20:18
did not wake up with their throats slit. This is a message to founders. If you partner with any of these people, they
00:20:23
will slit your throat and take your business wholesale. There is nothing to discuss here. Don't trust them. Use your
00:20:30
own models. >> I think I think it's less Sam and Open AI to be honest. I think that the
00:20:36
diversity of Open AI in terms of its revenue streams and specifically its consumer business may actually be its
00:20:43
savior in a relative value basis. Right now, OpenAI equity I think is more reasonably priced than entropic equity.
00:20:52
And the reason is not because of the quality of the models or the teams because they're both excellent teams,
00:20:57
but the reason is that OpenAI can fall back on a really healthy consumer business.
00:21:03
>> Sure, >> the difficulty that Anthropic is going to face is that I think what Sax said is
00:21:07
true that they have lost this fundamental trust about being able to stay within their sandbox. And if you
00:21:15
consistently demonstrate this tendency to learn and then to try to disrupt your host organism,
00:21:23
eventually you get sort of pigeonholed and you get cornered and people find ways to work around it. And so, look, I
00:21:31
sent that text that I had about the or that post that I had about uh our testing of of our harness on these
00:21:38
Chinese models to somebody well known in the industry. And he says, "Look, if you
00:21:43
also add some post training with all of the telemetry that you're going to get from the harness itself, he's like, I
00:21:50
suspect you'll find that it gets as good as Mythos." And and I thought, well, if
00:21:56
that's true, then why don't I just take GLM, control it entirely, soup to nuts, on my own hardware
00:22:02
inside of the United States with only US citizens that can touch it. It just seems like the brain dead obvious thing
00:22:08
to do, and it's much much cheaper. >> Yeah, and 100% correct. I will say if you're going to do this, you'll
00:22:17
eventually wind up rolling your own LLM. I mentioned a company, Abacus, goabacus.co, that we seated in our
00:22:24
accelerator. What they're doing now for HIPPA client people is they're actually giving you this go one box. They're
00:22:31
literally saying we're going to make your own model for you. So once you start this you you start with you know
00:22:37
claude and use their wrapper and everything or open AIS. Then you move on to the next step. The next step is I'm
00:22:42
going to use an open source model use my own try to find a harness etc. Where you
00:22:46
will eventually wind up is you're going to fork these models you're going to build your own. That is the end state on
00:22:51
prem on your own hardware. Don't trust anybody because there's too much at stake. You cannot risk your entire
00:22:57
business. You might as well be part of the vanguard and start investing here. You're going to slow down to speed up is
00:23:04
what's going to happen. Whether you're using 8090, Abacus, or any of these other solutions or rolling your own, you
00:23:08
must have AI sovereignty. You must have intelligent sovereignty or you're just giving your business over to your
00:23:14
competitors. The only company, the only company at scale that's ever respected the developer community in this way is
00:23:21
Apple. Apple took a very strategic approach to wanting to build an app store business to wanting to support
00:23:26
developers. And they explicitly told people, if you make something super obvious that you can build in a week or
00:23:31
two, it eventually might wind up in our basic collection of apps, the stock app in your iPhone, the notepad app in your
00:23:39
iPhone, those apps are incredibly basic. But if you looked at, you know, Robin Hood's uh or Google Finance and then you
00:23:46
looked at something like say Evernote back in the day, which was an advanced notetaker, it took 10 years for Notepad
00:23:52
to add the features that Evernote had 7 8 9 10 years ago. They specifically slow
00:23:57
their apps down to make them simple for users and not f with their ecosystem because they want to take the 30% tax.
00:24:06
That's your choice here. There is no 30% tax here when it comes to anthropic. There is no 30% tax equivalent with open
00:24:12
AI. >> The thing with Apple is that Apple was renting distribution. This is not renting distribution. This is where
00:24:19
you're renting intelligence and judgment. And so the problem that a company has is you can't rent the same
00:24:26
in this is why your point is actually right. I think it's for a different reason. You can't rent intelligence from
00:24:32
the same place that rents it to your competitor. >> Correct. >> You just can't. You can't. It's just
00:24:37
stupid. It >> It will overflow. it will go over the wall in your competitor's lap.
00:24:41
>> It's it becomes the lowest common denominator problem where you and your competitors now look exactly the same.
00:24:47
Why would you do that? And again, if you go back to that that thing that BCG identified, so many companies are
00:24:52
already teetering on a very difficult position where they cannot generate returns on their invested equity. And so
00:25:00
why would you then go and pay all this money so that you end up with the same answer as your competitor?
00:25:06
>> You can't do it. It's just not a it's not a choice. and Sachs, part of what's
00:25:09
going on here is the deflationary nature of technology. Every single one of these
00:25:14
tools is getting cheaper. Tokens are getting cheaper, Sachs. And if you look at Nvidia's role in all of this, they
00:25:21
have something called Neotron. You can try it if you uh use Perplexity. You can just pick the drop down menu. It has
00:25:27
deep thinking. You will not be able to tell the difference between Jensen Wong's open-source LLM sacks and Claude
00:25:35
for 95% of your searches. I guarantee you now why why why has Nvidia and Jensen downplayed their open- source
00:25:44
model until this moment? Why would he do that? Why would he never bring it up in
00:25:48
the all interview? Never bring it up because his top customers were uh very concerned from what I understand about
00:25:57
the fact that they had made so much project progress on their open source model. But suddenly after open AI
00:26:04
announced their jalapeno chips, after Anthropic started uh making chips, after AMD did successful uh projects with both
00:26:13
of these companies, after Elon said he's going to do his own fab, Nvidia's taken
00:26:17
the gloves off, David, they are going to own the whole stack. They are going to be talking about open source a whole
00:26:22
bunch. So maybe you could talk a little bit about Nvidia and their role in this as the open-source atscale full stack
00:26:30
provider. You get the hardware from them and you're gonna get a model that's competitive with open AIS for free and
00:26:38
all you have to do is use one of their hosting companies, Cory, whoever is buying Nvidia's, Colossus, etc. What are
00:26:43
your thoughts on Nvidia suddenly being willing to talk about their open source projects today? Well, look, I think part
00:26:50
of it is they needed time to make the offering compelling and you know, they're up against some pretty great AI
00:26:56
labs and I think some of it is just, hey, it takes time to train up these models. Now, one question is why is it
00:27:04
that Nvidia and Palunteer are partnering? Like what makes them natural partners? And I want to just explain
00:27:11
that. If you want to think about the AI stack for a minute at the most basic level, there's three layers to the
00:27:16
stack. It's basically the chips, it's the models, and then it's the applications. Well, right now we have in
00:27:24
the middle layer at the model layer of the stack, you've got two dominant companies. You've got Anthropic and Open
00:27:30
AAI. We know that Enthropic is around 60some billion of ARR. Open AAI is at 40 something billion of ARR. As far as we
00:27:37
know, no one else is really generating meaningful revenue at the model layer. So, we already have, let's say, an
00:27:43
emerging duopoly situation. We have Anthropic pushing for a regulatory capture agenda that would probably
00:27:51
enshrine that duopoly situation at a regulatory level because they're pushing for a safety agenda where Daria
00:27:58
explicitly says that these other models are not safe. You shouldn't have access to them. So you've kind of got that
00:28:04
situation. You got the market producing duopoly. You got the government now potentially leaning not to bust up the
00:28:10
duopoly but maybe to enforce it. So that's sort of the emerging situation at the model layer. And so if you're an
00:28:16
application at the top of the stack like Palunteer or you're a chip company at the bottom of the stack, that's the last
00:28:21
thing you want. You want a competitive model layer. Why? Because if you're an application, you don't want to be
00:28:26
beholden to one model provider, right? You want to have a choice. And if you're an enterprise, you want to have a choice
00:28:32
because you don't want have to give up all of your proprietary knowledge. And if you're a chip company, you you don't
00:28:37
want a monopsiny buyer situation where there's only one or two companies who can buy your chips and by the way,
00:28:42
they're producing their own. You want to have as diverse and healthy an ecosystem
00:28:47
as possible where there's lots of potential buyers for your chips. And if enterprises are rolling their own using
00:28:53
open models, that's kind of an ideal situation because now there's like a long tale of buyers. So I think really
00:29:00
the whole ecosystem in a way the chip companies, developers, applications, enterprises, everybody has an incentive
00:29:09
for a competitive layer of the stack at the model layer. Really the only companies who don't are anthropic and
00:29:17
open AI because obviously they want to dominate. They want to be a duopoly. And my view is, look, if you earn a monopoly
00:29:25
or duopoly in our system, we don't ban monopolies in the United States. We ban anti-competitive tactics, but if you
00:29:32
lawfully achieve a monopoly through amazing performance, we don't, you know, nationalize you or make you illegal, and
00:29:38
I think that's fine. However, the government, in my view, should do nothing to make monopoly or duopoly more
00:29:45
likely. They should make it harder for these companies to engage in monopoly tactics and they should do everything
00:29:51
they can to keep the model layer competitive because competition is what brings out the best and it's good for
00:29:57
the ecosystem and it ensures our civil liberties and consumer choice. So, >> and it's going to be amazing for pricing
00:30:04
sacks if you think about what this competition is about to do and you've been very vocal about this in your time
00:30:09
in Washington DC. We want to have a level playing field. We want to see massive competition. I'll give you your
00:30:14
flowers and your lay. You did a great job of now setting the table for this in 2026 and 2027 going forward by letting
00:30:22
people compete. The cost of tokens, Freeberg, is going to go down 90% a year for the next 3 years. You're going to be
00:30:28
able to buy a thousand times as many tokens that are more intelligent because you're going to have free options. The
00:30:35
price will be free or close to free for many of these and that could be incredibly disruptive. Yes, Dave.
00:30:42
>> Yeah. And I think people are going to again deploy their own hardware against
00:30:46
it. I think there's going to be a buying frenzy in the enterprise, not just with
00:30:49
the Neoclouds and the hyperscalers. I think the the enterprise is going to be a buyer. And when that happens, you do
00:30:56
the simple math and you don't want to have a dependency on server availability and cloud downtime and you don't want to
00:31:03
put these models on some third party cloud and you want to do stuff that's very cheap. Like a lot of people are
00:31:08
running I mean you guys do this. You're running day-to-day workflows for your enterprise and you realize, hey, I could
00:31:14
run these workflows on an open source model on a machine in my office and we don't need to be sending this stuff back
00:31:19
to some fancy scalable cloud service provider. >> Right? The industry spent so many years
00:31:23
convincing everybody to flip to the cloud and the realization may be that all this this idea of shared
00:31:31
infrastructure may not be the best idea in a world of intelligence >> and this is what I mean by like a
00:31:36
distributed spoke model because I do think it's not going to be all or none. And I think you're going to end up being
00:31:40
like 7020 20110 in how you're going to allocate your resources for model inference and running models. You're
00:31:47
going to probably be 70% in some big cloud. Maybe you'll do 20% local 10% you'll try other clouds. You know,
00:31:53
you'll kind of mix stuff up, but I don't think you're going to end up doing things the way you've been doing them
00:31:57
historically. You you'll very quickly realize that it's okay to waste tokens. It's okay to let your employees make
00:32:03
stupid apps that last for a couple weeks but burn through billions of tokens. If
00:32:09
it's running on your own hardware, then all you're paying for is the electricity
00:32:12
in your office or in your >> what I've been talking about here for like everyone's realizing. Yeah.
00:32:19
>> Like I'm totally fine, you know, uh for my kids to eat sugar if I don't have to
00:32:24
deal with them, you know, like go ahead. >> Have you guys noticed that like in the
00:32:30
last 3 days we've now seen other people trying to get their own lock in? Microsoft just announced a$ two and a
00:32:38
half billion dollar investment to stand up an FTE or Amazon is spending a billion dollars
00:32:43
>> forward deployed engineers for the people in the audience for deployed engineers has one open AI has one.
00:32:49
>> Yeah. And when they come knocking Jam they're knocking like hey can I send my
00:32:52
engineers to study your business and put it into my model? I mean people are going to be slamming the door on these.
00:32:56
It's like getting a Jehovah's Witness at your door. Like no I don't want to be part of your cult. I want to own this.
00:33:01
>> And that's what I've been saying. Just let me make one quick point and I'll hand it to you for free break. This is
00:33:05
what I've been saying with when I when I was going on my open claw which now is like Hermes and some other products.
00:33:11
Everybody in your organization is going to have a max studio or a Dell with a massive amount of RAM and you're going
00:33:18
to spend 10 $20,000 per employee on local compute so that they can token max to a [ __ ] maxing level. Who cares what
00:33:26
they do on their local computer? Who cares? Let them rip. And then you're going to give them a laptop that
00:33:31
connects to it and it syncs so you can control it. It's literally going to be a server per individual in your company.
00:33:38
That's the way to model this in your brain. Everybody has their own language model that they're crafting 100% of the
00:33:44
time as they work and it's all local so you don't have data leaks. Go ahead, Freeberg. I'll give you the final word.
00:33:50
I mean, I think one of the things that this reveals quite clearly is that there is no buttered slippery slide to job
00:33:59
loss. >> I think everyone is realizing that, you know, >> I was wondering where you were going
00:34:03
with that buttering it up. So is Truman. >> But I think, you know, the idea that
00:34:08
everyone had in their head two years ago, the narrative that was formed and everyone clutched on to, and by the way,
00:34:14
you will not see the media reverse on this narrative. I've realized the importance when when
00:34:22
>> media >> if you're a reporter or you're a journalist >> and you come up with some story that
00:34:29
says something that's happening or is going to happen or has happened or is about the future, you destroy your own
00:34:35
credibility if the narrative shifts. You cannot ever let go of the narrative. And
00:34:40
I think this is one of the things that that we can kind of acknowledge is going on with this job loss narrative problem
00:34:46
is that even as all the data comes out as all of the reconfiguring of how enterprises are using AI as it's
00:34:53
revealing to all of them that they're actually not just going to cut cost but they're going to grow revenue and it's
00:34:58
going to be a kind of clunky way of getting there. It's not going to happen overnight. It's not this slippery slope
00:35:02
to job loss to nealism. Everyone's gonna kind of wake up and be like, "Wait, this
00:35:07
reality that we all thought we were living in is not really the reality that we are living in. We're living in a
00:35:12
reality where AI is clunky. It takes some putting together. It's a little more complicated than we thought. It's
00:35:19
definitely going to deliver value, but it's not about just turning off all the jobs and letting the genius AI solve all
00:35:24
my enterprise problems and scale me into infinity without humans." And I think that that's a big kind of narrative
00:35:29
shift. And you will not see the media accept that their narrative is wrong. Because as soon as they have to
00:35:35
acknowledge that they were wrong in what they were saying about job loss and all
00:35:39
the other senators and people that are proclaiming job loss, job loss, job loss, by the way, the reason they're
00:35:44
making that proclamation is so that they can step in and control AI and they can
00:35:48
drive their systems of socialism, which is what they're all looking to deploy. But if they had to come in and say,
00:35:53
"Look, the data doesn't map to the narrative, their credibility is destroyed." So they'll double down on
00:35:58
it. and they'll double down on it. And I'm telling everyone that's listening, look at the data. There is no job loss
00:36:05
with AI. It is an absolute scam to tell the world that AI is taking away jobs and destroying jobs and the world is
00:36:12
shifting. It is clunky. It is valuable. It is going to take some time and it is going to create far more jobs than it is
00:36:18
destroying. >> We're a bunch of monkeys. We've been given this new tool. We're going to
00:36:22
solve more problems with the new tool. There will be job displacement. And the problem is people do not understand the
00:36:30
nuance between these two terms. Certain jobs will be retired and they're going to be retired at a faster rate than you
00:36:35
can imagine, but other jobs will happen. And you pointed this out, Shimoth. Hey,
00:36:40
it's the it's the most I don't know if you did it here or on your personal channel on YouTube, but you know, it's
00:36:45
the most empowering tool ever. If you can What jobs are getting replaced? >> I I challenge very simpl,
00:36:52
>> but you're saying some of them are going to go away faster than you can see. What? What are those jobs that are going
00:36:55
to go away faster than you can see? Cuz I'm just I'm still not seeing the categories.
00:36:59
>> Totally. Customer support jobs. Those are going to go away very quickly. >> Where? Where? You're just making [ __ ] up
00:37:04
cuz it's not I see it in enterprises all the time. >> Where do you see it? Show me the
00:37:10
customer support enterprise shutdown that's happening. Just show it to me. >> Is it in the room with us right now?
00:37:15
Like where is it? I will literally, you know, I'll I'll do that research and and
00:37:19
I'll give it to you. >> I'll give you some research. Hold on. Can I can I provide some research?
00:37:24
He asked me a question. Just let me finish my answer and then you can take it. You asked me the question. Customer
00:37:29
service jobs are going to go away because consumers prefer to talk to the AI and it's perfect at that. Entrylevel
00:37:36
jobs around data entry around business process outsourcing those are going to go away. So will driving cabs and those
00:37:42
kind of jobs. Those ones are obvious lowhanging fruit. Just like we had the typing pool and messengers and other
00:37:49
products go away when we had word processors on every desk. Those jobs will be displaced 100%. And those jobs
00:37:55
>> year we're on year 20 of cab drivers going away, dude. Like when Uber came out, everyone said all the driving jobs
00:38:00
were going to go away. It's all over. >> Well, no. Whimo is stuck at 3,000. Tesla's stuck at like 30 cars. But
00:38:08
that's all going to change in the very near future. >> Everything you're saying is perspective.
00:38:12
This is my point. You are still being perspective in the future with everything you're saying. And every
00:38:18
point that we turn, every time we turn a page, it's like, wait, do you happen to
00:38:22
think self-driving is real? Is that your premise? Is that self-driving is not going to get rid of cab drivers? Is that
00:38:27
your premise? >> I think it's real, but like like I think it's real for sure. >> Do you think it will get rid of cab
00:38:31
drivers? You're arguing as if like I'm wrong here, but I think that every day I get in the car every day and my car
00:38:38
drives me and I get on my phone and I do work. >> Okay. But do you think cabs and cities
00:38:42
are not going to be self-driving? Do you think zipline's not going to deliver? Bailey thing again. Jal,
00:38:48
>> it's not Mont belly. I see this from my investments. I have a company auto lane
00:38:52
that is doing this right now. >> But just go to customer support again cuz you've been saying Yeah.
00:38:56
>> I literally we've had this debate verbatim and >> displacement not job loss is is my only
00:39:01
point. >> Yeah. Yeah. Yeah. Yeah. What Freeberg is pointing out is that whenever anyone
00:39:05
pushes back on the fact that you don't have any data to support this in the present, you say you're talking about
00:39:10
the future. Okay, fine. It's a prediction. You know, there's no data to support it in the present. That's fine.
00:39:16
You're saying it's gonna happen in the future? >> No, no. I'll give you the data point.
00:39:19
>> You can dismiss I want. I will. Yes. If you talk to Uber and Whimo, in the cities where Whimo is present and has
00:39:24
gotten past a couple hundred cars, they've stopped recruiting drivers. Drivers are either static or going down
00:39:30
in those markets. So that is absolute evidence that this is happening and the roll out is going to be fast and
00:39:35
furious. So it is not a future prediction. You can talk to the CEO of Whimo. You can talk to the CEO of Lyft
00:39:39
and Uber and they will tell you this explicitly. Market interview Darren. He said jobs were
00:39:45
increasing at Uber. >> Yeah, cuz you're delivering more like those because productivity is going up
00:39:50
with with delivery. Productivity goes up means people drive more% lying. You're being dis you're being last time we no
00:39:57
the last time we disingenuous in markets where Whimo is in markets where Whimo is you're you're there might
00:40:05
be other markets where things are growing because they don't allow self-driving or they haven't gotten
00:40:09
there yet. In markets where Whimo has hit critical mass, the number of human drivers is going down. This is just a
00:40:14
fact. Stop recruiting in some narrow arbitrarily defined subset of their jobs. You're pointing to flatness.
00:40:22
>> Meanwhile, Dar told us that hiring a cost of the company has gone up in the face of these drivers.
00:40:28
>> No, no, you're wrong. Okay. Anyway, we we'll agree to disagree here. >> Let me give you some actual data. Can I
00:40:34
provide us this study? So, RAMP and Rellio Labs just released a new study in the past week.
00:40:39
>> Yeah, that was great. >> And it was an actual study of over 21,000 firms in the US and they looked
00:40:46
at their payroll data combined with their spending on AI. And what they saw is that firms that spent the most on AI
00:40:55
actually grew the fastest and they tended to grow their headcount roughly 10% in the two years following the
00:41:03
adoption of AI. And entry level headcount rose even faster, it grew at 12%. So all this stuff about how entry-
00:41:11
level headcount's going to get wiped out, not true. The more firms adopted AI, the more hiring they did. At least
00:41:17
that was a correlation. Obviously, they can't prove causation, but that was the correlation. And then companies that
00:41:24
were not highintensity adopters of AI, they were either didn't adopt or they were low intensity adopters, they just
00:41:31
saw flatness in their headcount. So, no one is really showing a trend here of job loss. What they're showing is that
00:41:38
the more you adopt AI, those companies also tend to grow headcount. So I you know there's just no data to support
00:41:45
this idea that in the present AI is causing job loss. >> Job displacement will happen. Jobs will
00:41:52
increase because of AI. Sovereignty will increase and people will make more money. Corporations will be more will
00:41:57
have more earnings. But there will be jobs being retired. And I think if you look at translators, if you look at
00:42:05
telephone operators, if you look at bill collectors, if you look at Whimo drivers, these very simple tasks are
00:42:11
being automated. How do I know that? Because I invest in startups that are raising hundreds of millions of dollars
00:42:18
to replace these jobs with AI. We had the pilot uh we we had people building the perfect pilot, the perfect driver,
00:42:26
the perfect customer support rep. That is the point of AI. Every single sales pitch of these AI companies is to retire
00:42:33
these jobs that are not great jobs. There is no way in God's green earth that humans will be at Amazon sorting
00:42:41
packages. There's not. It's not going to happen. All those factory jobs are being
00:42:45
replaced. Elon is explicitly building Optimus for this purpose. And when I say it's future, I have always contended
00:42:51
it's future. Sachs, you cannot use your silly debate techniques to say that I'm using some crazy thing. It is happening
00:42:57
as we speak. We're soaking in it. This is the transition period. What I'll tell you is that in every
00:43:04
environment in which 8090 operates, every customer, I don't think we've seen layoffs. Not a single one.
00:43:11
>> Yeah. In this in this RAM card study, they said, quote, "Gains emerge gradually and are broad across roles,
00:43:18
including engineering, sales, administration, and customer service. So, they're not seeing like one category
00:43:23
of jobs getting wiped out." By the way, it could happen in the future. I mean, none of us can say for sure what's going
00:43:29
to happen in the future, but if you're looking for data in the present, we're not seeing it yet.
00:43:33
>> I think there's going to be a strong human alpha. You know, like when vending
00:43:37
machines came along, it's not like bartenders went away. Like I like going to a bar talking to the bartender and
00:43:44
they make a great drink. And those bartenders make better money today than they did 20 years ago. I don't and and
00:43:49
maybe the vending machine replacement model isn't the right one in that case, but I I think that there's a lot of
00:43:55
value in human interaction that's going to get a premium in the future because as more stuff gets automated, I want to
00:44:01
have an actual driver when I go to Las Vegas. I don't want to go in a Whimo. I want to have a great driver in a nice
00:44:08
car that takes me to the dinners, drives me around New York City, same thing. I don't want to go in a Whimo, per se. I
00:44:13
want to have, you know, people that can do things that could otherwise be automated. I'll pay a huge premium. I
00:44:18
could sit in a freaking massage chair. Why do I go pay someone $200 for an hour massage? You know, I think there's
00:44:23
there's an aspect to the alpha of we could cut that one. But uh but I do think there's a premium to
00:44:31
humans and I don't think that um and I actually to humans >> no the count JL this is really important. I think
00:44:39
that the counternarrative to automation generally is that we're going to realize
00:44:44
the importance of human interaction and humans in the loop and we're going to pay a premium for it.
00:44:50
>> Well, do you remember like a >> I don't believe I don't believe in AGI and the singularity happening next year.
00:44:55
I'm just not there. >> Do you remember like a year ago when CLA said they're going to replace their
00:45:00
whole customer service department with AI and they did they did a lot of hype. It was peak RF farming and well lo and
00:45:08
behold they flipped the decision back after a year. They said that from a brand perspective, a company
00:45:13
perspective, it's just so critical that you are clear to your customer that there will always
00:45:18
be a human if if you want it. >> Human in the loop will be the premium and we will all realize in the next year
00:45:24
and a half the importance of human in the loop even in the context of AI and automation working.
00:45:29
>> Human in the loop is going to be more valuable and then you're going to pay a
00:45:32
premium for human in the loop. There's so many CEOs who want to tap into the media's hype cycle. This is the thing is
00:45:38
that whenever the media gets a hold of one of these narratives, there's so many CEOs who are not like Karp, right, who
00:45:44
just are not original and they don't really have anything important to say. So, the only way for them to get
00:45:49
publicity is to glom onto that press cycle and try to use it. And then they'll chime in with, you know, how
00:45:56
they're going to do crazy things to eliminate their whole customer sport department and then lo and behold, it
00:46:00
turns out just to be total totally fake. I think the ramp data is directionally accurate. In our experience, I would say
00:46:06
the same thing. I think people that use AI tend to grow faster. They tend to hire more. They tend to make more%.
00:46:12
>> Yeah, 100%. >> By the way, I actually agree with 100% agreement on that. >> From a career standpoint, I would not
00:46:17
want to be a level one customer support rep because I do think that like if all you're doing is like answering forgotten
00:46:25
password requests, that's not a great place to be. However, those jobs were basically outsourced to the Philippines
00:46:31
and places like that a long time ago. I mean, most enterprises are not doing kind of that level zero, level one
00:46:38
customer support in the US. They do have customer support centers, but they're doing like level two, level three.
00:46:43
They're doing the escalations, the higher value stuff that's going to be a lot harder to replace with AI.
00:46:48
>> Totally. I do kind of worry about well what happens in some of these other countries that are doing the super entry
00:46:53
level stuff because I think that is job loss but I don't know if it'll be massive but that's where the risk is. I
00:47:00
mean frankly it's not in the US. >> This is um so maybe we're getting to some consensus here.
00:47:07
>> Job displacement versus job loss. Here is Brett Atcock sharing what they're doing at Figure. They had an 8-hour
00:47:15
challenge to have this uh robot they're building sort packages and they did it for 200 hours. That's today. This is the
00:47:23
least good it will ever be. This is the worst it's going to ever be. These things are
00:47:29
getting really good. I spent time with Elon. I've looked at Optimus. I can tell you Optimus is better than what Figure
00:47:34
is doing. What Optimus is going to do when Bezos and Andy Jasse unleash Optimus inside of the Amazon factories
00:47:42
is it's going to get rid of every, let me state this very clearly, every single package sorting, every single package
00:47:48
delivery will not be done by humans in 10 years. And it's starting today. Just like the idea that you would put
00:47:55
together some of this consumer electronic stuff in factories with humans has gone away over time. and
00:48:02
other countries. I think you're making a very good point here, Saxs, are going to
00:48:05
experience it first. In America, knowledge workers, the entrepreneurial spirit. I think we're going to see a
00:48:10
Cambrian explosion in startups, which means we're going to be the ones who benefit from it. But if you're looking
00:48:15
at those other countries, they'll see it acutely first. We're just going to see it here with drivers, and that's 5 10
00:48:20
million people, like we saw it with cashiers. The idea of going to a cashier at a fast food restaurant, which I know
00:48:25
you guys don't go to all that often, where Starbucks, the idea that you would interact with a cashier has been going
00:48:30
away slowly because it's just consumers don't even want it. We'll agree to disagree here to a certain extent. Maybe
00:48:37
we may >> Look, I went to McDonald I went to McDonald's recently and I ordered from
00:48:40
one of those monitors. Zero. >> What did you order? You filet a fish guy? I'm a flay of fish guy. What'd you
00:48:44
get? What's your >> I'm a Big Mac. I like a Big Mac. >> You like a Big Mac? What about you, Sax?
00:48:49
What? What's your order? What's your Did you get a McFlurry? They had a thing called a homegrown
00:48:54
burger which was some new I think it was kind of like more like an In-N-Out type
00:48:58
burger situation they were testing. >> Smash Patty. >> Yeah, it was good. >> Anyway, my point is just the automation
00:49:04
obviously they didn't have a cashier there already, but there's zero robots involved. Similarly, there's already a
00:49:09
lot of automation going on in those Amazon warehouses. I don't know if you've ever been to like a FedEx
00:49:14
packaging or routing depot. It's all conveyor belts and you know >> Yeah, robots pushing it and directing.
00:49:20
But it's not it's not humanoid robots. There's already tons and tons of automation at these places.
00:49:24
>> The last but this last mile part is the one I think that is going to shock people. When you have an optimist
00:49:30
walking the Tesla self-driving car and then they walk it to your porch, ring your doorbell and knock on your door and
00:49:37
put it on your front step porch. That's where this is going to get super interesting.
00:49:40
>> I think it's going to lead to a boom because think about like just something like building a house. Okay. You know
00:49:45
how long it takes to build a house >> in California? Three years. Yeah. Yeah. Yeah. Multiple years now
00:49:52
>> in >> China one day. >> There you go. >> But imagine if you have like a like a
00:49:58
construction crew and I don't know it's like 20 humans and 50 robots or something. The humans have to tell the
00:50:04
robots what to do. I don't >> supervisors. >> I think it's going to take a long long
00:50:09
time for the robots to be able to do that super high skill work. But they can be given tasks
00:50:15
>> and then gradually they can do. So maybe it only takes a year to build a house
00:50:18
instead of two or three and that's the solution to our housing crisis. >> 100%. All right, listen.
00:50:24
>> In a related story, by comedant Howard Lutnik's decree, subject Dario Amade will be permitted to bring his powerful
00:50:33
new model Fable 5 to market for the glory of the realm and at the pleasure of his majesty Lord Trump. Remember two
00:50:39
weeks ago the US government put export restrictions on Fable and Mythos and the Iliad and the Odyssey and whatever else
00:50:44
they're doing over there at Anthropic and Anthropic expanding Mythos uh to 50 unauthorized entities according to
00:50:53
sources that included SK Telecom which allegedly perhaps maybe has li ties to China on June 30th. Howy in the commerce
00:51:02
department. That's my guy, Howard Lutnick, my favorite in the administration. No disrespect to you,
00:51:06
Saxs. Um, he lifted controls on Fable 5. Mythos 5 was restored to US customers this past week, June 26th. Many
00:51:14
suspected this would happen because Anthropic replaced Daario as their lead negotiator. Here is the Palace Intrigue
00:51:20
I was referencing earlier. Co-founder Tom Brown, whoever that is, took the spot and appears to be getting along
00:51:26
better with the Trump organization. For example, he proactively genulected showed the tweet to Lutnik's tweet
00:51:33
announcing the export restricting restrictions had been lifted. Tom Brown, thanks for your partnership on this
00:51:39
secretary explanation point. Look at that smile. Tom Brown coming in to save the day. In a postmortem blog post,
00:51:45
Anthropic said the jailbreak that triggered the shutdown was not unique to Claude. Remember Andy Jasse reported to
00:51:51
the Commerce Department what was going on with this. Okay, Saxs, former AISAR. What's going on? Give us the palace
00:51:58
intrigue, please. >> Well, look, the export control letter was taken down after 2 weeks. I think
00:52:04
that this was a highly unusual circumstance, and it was brought about by three things, three conditions, and
00:52:11
you really needed all three in order for this to happen. Number one, you had Daario running around for months saying
00:52:17
that he had created a cyber weapon. He almost boasted about it. It was mythos. Okay, so that was number one. Number two
00:52:23
is you have a trusted partner of Anthropic which is Amazon which did its testing of Fable which was Mythos with
00:52:31
guardrails and they reported that the guardrails failed. So therefore Fable was according to Daario a cyber weapon.
00:52:38
So that was important fact number two. And then fact number three is that when confronted with this information, Daario
00:52:48
basically refused to roll back Fable until the jailbreak could be fixed. Or at least that's what he communicated to
00:52:55
the administration and that's what the administration heard. So I think if you change any one of those three facts,
00:53:03
this would have ended up not causing the government to basically send that that letter to Anthropic. I think if Daario
00:53:10
hadn't primed officials to see Mythos as a cyber weapon, it wouldn't have happened. If their trusted partner
00:53:17
Amazon hadn't reported the jailbreak, wouldn't have happened. And if Daario hadn't refused to take action when he
00:53:23
did, then this wouldn't have happened. I think this was sort of a unique circumstance. I guess what I'm trying to
00:53:28
say is that a lot of people are reading a lot into this letter. I know that foreign companies or foreign actors or
00:53:36
allies and partners of the US are wondering does this mean that our ability to access US technologies is
00:53:43
going to be limited? I don't think it does. I think that the administration reacted with the tools at its disposal.
00:53:50
And I think that what the president cares about is what he's always said, which is he wants to be pro- innovation.
00:53:57
He wants to be pro export. He wants to be pro infrastructure. He wants to support American companies in the AI
00:54:02
race and I don't think people should overextrapulate based on what just happened over the past few weeks. I
00:54:08
think highly particular to this fact pattern. >> We a lot of discussion of exports. Let's
00:54:12
talk about imports for a reason. It seems to me >> that importing Chinese models when we in
00:54:18
fact have open- source models here from Nvidia and other players. Why have you not or in your during your time did you
00:54:24
not advise or talk about here blocking access to import? We will not allow the importation of Huawei products into the
00:54:32
United States or self-driving technology from China. Why are we allowing open source models? Why don't we put an
00:54:38
edict? And why didn't you suggest that when you were thesar of AI? Why didn't we stop Kimmy and Deepseek from coming
00:54:45
into this country in order to drive our open source technology and to keep the leakage
00:54:52
from those model sacks? >> Well, if you is that a concern of yours? Well, number one is that once a model is
00:54:59
open- sourced, it stops being Chinese in a way, right? Because you can now take that model, you can fork it, you can
00:55:06
create your own version, you run it in an American data center on your own hardware. There's no packets going back
00:55:12
to China. There's no data leakage going back to China. In some sense, they've made a contribution to the open source
00:55:17
community. And then people can take it from there. Now, should you still exercise caution? Absolutely. because
00:55:24
you have to make sure that the model is safe. It doesn't contain back doors. I mean, this is a relatively new surface
00:55:31
area for cyber security. So, look, you should obviously be cautious. But the bottom line is that when an American
00:55:39
company takes an open source model and converts it and runs it itself, it is now theirs.
00:55:44
>> But it builds consensus around that model and makes that model stronger by more people participating in it, which
00:55:49
was your argument for allowing people to get on the hardware stack. Well, but then that I think that leads
00:55:56
me to argument number two, which is if you were to do something like ban open source in the United States, you'll put
00:56:02
the United States on an island. The rest of the world is not going to follow suit. The rest of the world wants to use
00:56:07
open models because of the advantages they offer. They're cheaper. They're more customizable. They offer more
00:56:13
control. And again, they stop being Chinese models once you can take them and adapt them and run them on your own
00:56:21
hardware. So the rest of the world's not going to stop using these models just because we do. And what we will do then
00:56:28
is subject American enterprises to a token tax. You're going to end up paying for closed models in the US.
00:56:35
>> No, but we have open source models now. Nvidia has theirs, for example. There
00:56:39
are other ones. So if there are open source models here that are equally competitive,
00:56:43
>> would you then support an import control so that more people in America put their
00:56:48
effort into American open source models? Well, I've been I've one of the things I've been saying for
00:56:52
>> and Google's open source models. Obviously, >> one of the things I've been saying for
00:56:54
over a year is that we need more strong offerings in open source from the United
00:56:59
States. So, I hope we win the open category of models just like I do the closed. I mean, I hope America wins the
00:57:06
AI race in both open and closed models. I don't really want anyone being forced to use models they don't want. But look,
00:57:13
if American models, open models are better than Chinese open models, they will win and people will want to adopt
00:57:19
them. So if what you're saying is true that our open models are better or getting better or going to surpass the
00:57:25
Chinese ones, leave that to the market to decide. And just one last point on this is, you know, I'm not against
00:57:31
limiting the import of certain Chinese products in the US. For example, I don't think we allow Chinese connected cars.
00:57:37
>> We don't >> I think we should think twice about whether we allow, you know, Chinese
00:57:42
robots into the US. I mean, >> yeah, we wouldn't want to have like a like a fifth column of,
00:57:48
>> you know, of backdoor Chinese robots or something in the US potentially. I'm just brainstorming here. Okay. So, I'm
00:57:53
not against limiting certain things, but we should understand that when we do that, we are inviting retaliation by
00:57:59
them as well and we are in a trade relationship with them. We still need things from them. We still need rare
00:58:04
earths, things like that. One day we won't hopefully. I do think we should try to be as independent and autonomous
00:58:11
as possible, but while we do, we have to think very carefully about the larger trade relationship. Yeah, Trump banned
00:58:19
the import of Chinese drones. It's a signature part of the legislation. So, um, all right. So, a little housekeeping
00:58:25
here. All-in summit is coming back September 13th, 14th and 15th. Fifth year of the summit. Keep raising the
00:58:30
bar. Man, if you knew the speakers that I knew that Freeberg's lined up, your brain would explode this year. The
00:58:37
nights are as good as the days because let's face it, Freeberg likes to party. He's been known to have a cocktail or
00:58:43
two. We're doing the biggest poker tournament we've ever done at the summit this year. Gaming will happen at the
00:58:47
welcome party and we're going to have a full casino night at an exclusive mansion in LA. Ooh la la. Applications
00:58:54
open now. Allin.com. Don't miss our biggest event of the year. I think we have 12 trillion in market cap books so
00:59:01
far. It's probably going to double. Okay, next topic. Topic number three. It's a hot scot this summer. The
00:59:06
justices released 11 decisions in the final few days of the term. Let's discuss the top three right now.
00:59:12
Birthright citizenship case was Trump versus Barbara. On the first day of his second term, Trump, as you know, signed
00:59:19
an EO ending automatic citizenship for children born to illegal immigrants or those on temporary visas. Uh this is
00:59:26
codified, I guess, in the 14th amendment. Court struck it down 63, maybe 54. There's some nuance here.
00:59:32
255,000 children are born every year to non-citizen parents. Chief Justice Roberts wrote, "The majority, joined by
00:59:39
all three liberal members." The court is still a bit uh polarized citizenship then and now was the right to have
00:59:46
rights. We keep that promise today according to Chief Justice's majority. Kavanaaugh carved out a lane for
00:59:53
Congress to tighten up on birthright citizenship. That's still consistent with the 14th Amendment. Uh, as you
00:59:58
know, the administration has been pointing out, and you've watched Fox News, you've heard this a million times.
01:00:03
If you watch CNN and Not All In, you're going to maybe have heard of this once or twice. Birth tourism. A Chinese
01:00:10
national, Dong Wan Lee, plead guilty in 2019 to a scheme to help pregnant Chinese women enter the US under false
01:00:16
pretenses so their babies would be granted US citizenship. She pled guilty to one count of conspiracy to commit
01:00:23
immigration fraud and one count of visa fraud. So we we handled that through the
01:00:26
legal system. She served 10 months in prison. Trump's response to the loss. Quote, "Congress should start today to
01:00:35
work on ending birthright citizenship. They will have my complete and total support for Trump. Feels like a loss for
01:00:42
Trump objectively, but he's going to keep fighting this one." Close decision. Uh if you ask
01:00:48
Steven Miller, he doesn't like immigrants like you three dudes. But I would let all three of you in. Again,
01:00:54
>> that's not his position. >> Okay. Well, you can explain. I know he doesn't like immigrants.
01:00:58
>> We're legal immigrants. We're legal immigrants. >> Yes. Well, but >> you understand the difference, right?
01:01:03
>> Well, birthright would also be legal according to the Constitution. >> Yeah. But that's the whole point is it
01:01:08
doesn't really make sense that let's say an illegal immigrant runs across the border, is like 9 months pregnant, plops
01:01:14
out a baby, and all of a sudden they're an American citizen. How does that make sense? Uh, what about somebody who's
01:01:19
been here for 20 years, Sachs, worked really hard, they're illegal, and they have a baby. Should the baby kicked out
01:01:25
and that baby not be a citizen? You're, let's say it's somebody's nanny, somebody's gardener. We have 20 million,
01:01:30
30 million illegal immigrants we've let into this country. We wave them in across many different Republican and
01:01:35
Democratic terms over the last 30 years. What do you say to that person who has a
01:01:40
baby? Should they be deported? Should they be given citizenship? >> Well, look, the question here is what
01:01:44
the Constitution says. And my view is that the original purpose and understanding of the 14th amendment was
01:01:50
to make sure that the children of freed slaves would have citizenship rights. That was the purpose of it. That it's
01:01:57
obvious. And I don't think it speaks to the situations you're talking about and Congress should just make the law about
01:02:03
those situations. But now Congress cannot make the law because the Supreme Court has ruled that citizenship is
01:02:11
determined by birthright. So in that case, My question was somebody who's been here 20 years as an illegal
01:02:17
immigrant underrange. >> I'm saying Congress should >> What do you think? I'm asking you your
01:02:22
personal opinion. Should they get >> I mean, look, that's you're talking about an edge case. It's complicated.
01:02:26
But the point is that we all know why the 14th Amendment was ratified. It was to protect freed slaves, obviously. And
01:02:35
now we're in territory that's like very different. But the Supreme Court has decided that if you're born here for any
01:02:41
reason whatsoever, even if you're an illegal, even if you weren't supposed to be in the country, now your children are
01:02:46
American citizens. By the way, I don't think any country in the world has this policy. It's kind of a crazy policy. So,
01:02:52
I agree with Stephen Miller about this. >> I'm not sure it's an edge case. I think
01:02:54
I think the majority case is people who've been here illegally for for decades. And I think the edge case, the
01:03:00
one you proposed is the edge case, like a pregnant woman running across the border.
01:03:03
>> Let Congress decide that. I don't think the Constitution ever spoke to this. What's your personal opinion on this,
01:03:07
Shimat? Should the person who's been here for 10, 20 years, paid their taxes, but they're illegal, and they have a
01:03:12
child, should that child be a citizen or not? What's Shmat's personal opinion here?
01:03:18
>> I haven't thought about this. >> Great. Okay. It's uh I mean, that's the key to the uh I think what we're talking
01:03:24
about here. Free. >> No, no, no, no. Because J, there's a big difference between making something
01:03:29
constitutional and then doing it via a law of Congress. And what Supreme Court has said here is that Congress does not
01:03:36
get to make the decision. The majority of the American people don't get to make the decision. The Constitution requires
01:03:42
a certain point of view and there's no way to change that without a constitutional amendment, which will
01:03:47
never happen. So, we're kind of now stuck with this position. Hm. Well, if you read the language and someone posted
01:03:53
this, this was the debate that when the 14th amendment was being debated, Senator Jacob Howard, I'll give credit
01:04:00
to Guyer Capital for pointing this out. There's a transcript of the debate that took place and it says um every person
01:04:07
born within the limits of the United States and subject to their jurisdiction is by virtue of natural law, national
01:04:12
law a citizen of the United States. This will not of course include persons born
01:04:18
in the United States who are foreigners, aliens who belong to the families of ambassadors or foreign ministers
01:04:23
accredited to the government of the United States, but will include every other class of persons. Because if you
01:04:29
go to the history in 1868, there was this Dread Scott decision which denied citizenship to the children of slaves.
01:04:39
And the intention at the time was to ensure citizenship for newly freed slaves and their descendants. That was
01:04:44
the original intention of the 14th amendment at the time. So this goes back to the great debate that happens at the
01:04:50
Supreme Court all the time, which is the textual or the intentional reading of the Constitution or its amendments. The
01:04:58
textual reading is you look strictly at the words, the plain literal meaning of those words. And then the intentional
01:05:05
reading is what was intended at the time by the framers of the constitution or the amendment that may be the subject of
01:05:12
the debate. And so I think that's ultimately what this comes down to. If you think about the intention at the
01:05:18
time, it was really built around free slaves. There wasn't a consideration at that time of having this massive
01:05:24
movement of immigration that was taking place or that has taken place since. that wasn't even in the kind of realm of
01:05:30
consciousness or consideration when this amendment was being passed. So the intentionalism kind of argument would be
01:05:37
hey wait a second you know this wasn't really what they intended but the textual representation is what it is and
01:05:42
that's what the ruling kind of led to in this particular case. my personal opinion, if you're asking it,
01:05:47
>> I'm trying to get personal opinions here. >> I do think birthright citizenship um
01:05:51
should be endowed to the children of uh legal residents of the United States, whether they're a citizen or not. So,
01:05:59
yeah, I think that's critical. So, if you're a visitor, if you get on an airplane and fly here for a weekend on a
01:06:05
vacation and you have a visa or you have, >> you know, temporary visitor status to
01:06:09
come. >> Yeah. or or you have a temporary visitor status to come to the United States to
01:06:15
visit, you are not a resident of the United States. So, your children should not become citizens of the United States
01:06:21
unless you're a resident. And remember, we don't just have like citizens and non-citizens. We also have resident
01:06:27
status in this country. And a resident status is a green card. I mean, for years, I was on a green card. Three of
01:06:32
us were immigrants. We all went through this process. Once you get resident status, you're now a permanent resident
01:06:37
of the United States. You haven't been given citizenship rights yet. But if you have a child, I think it's reasonable
01:06:42
for your child to have citizenship rights if you're a legal resident rather than a legal visitor or illegal visitor
01:06:49
to the United States. That would be my personal assessment. I you know, it would be great for that to be the law,
01:06:53
but obviously this is now kind of, >> you know, gone explain that really well. I didn't even know all of this. I'm I'm
01:07:00
on your side. I agree. >> Yeah. I mean, I I agree with that. You're saying in addition, people
01:07:05
shouldn't be able to do the tourism thing. We should handle that. Um, I do think we should have an exception here
01:07:10
for people who have been noncriminals who are here illegally because we as America, Republicans
01:07:18
especially, uh, during the time when they were pro free trade, NAFTA and really waved in a large percentage of
01:07:25
the immigrants who here, obviously Biden did that as well. I think we have a moral obligation to those people since
01:07:30
we waved them in and we waved them in to work at our businesses because we wanted
01:07:35
to pay under minimum wage to them. those people suffered here in our country to take those jobs to make us richer, to
01:07:41
make our country richer. We have a moral and ethical obligation to them and to their children. We should give them a
01:07:46
path to citizenship unless they're criminals and their children should be Americans. Since they came here to live
01:07:54
the American dream, whether it was illegal or not, we waved them in. That was our mistake. And now that we've
01:07:59
closed the border, we should make amends to those 20 million people. That's my personal opinion. I don't think we have
01:08:06
need more time to think about it and Saxur. >> Well, my my point is we're all just
01:08:11
making arguments about what the law should be and these are the kinds of arguments that Congress should be making
01:08:16
and deliberating over >> and these are the edge cases. These are the nuances that should be handled in a
01:08:22
citizenship law. And what the Supreme Court has done here is remove that space for Congress to legislate. And I think
01:08:31
Congress needs that space because there are these edge cases and instead it's interpreted the 14th amendment in a way
01:08:37
that I don't think is consistent with why it was ratified in the first place. And Freeberg is totally right about
01:08:43
about Dread Scott which was one of the most awful probably the most awful Supreme Court ruling of all time. It
01:08:51
came out in 1857 and it ruled that black people whether enslaved or free were not US citizens
01:08:58
and therefore they had no right to sue in federal courts. And the chief justice Roger Tony wrote the decision I mean
01:09:04
it's extremely racist. He wrote that black people had quote no rights which the white man was bound to respect. And
01:09:12
that decision furthermore ruled that Congress could not ban slavery in the new western territories which meant that
01:09:18
it struck down the Missouri compromise which was the law that had kept the peace between the free and slave states
01:09:24
for decades. And that is one of the direct things that then triggered the civil war 3 years later. So it was this
01:09:30
historically bad decision that I guess then the 14th amendment tried to correct by saying that no
01:09:37
you know freed slaves and their children would be >> totally different context than someone
01:09:43
visiting from China for the weekend and having a baby. >> Yeah, which we obviously that one is an
01:09:48
easy one to adjudicate. Chimath, I'm curious feels like as a society we have now because of the closure of the border
01:09:55
give Trump a lot of credit for that. It was easily done, which means it could have been done at any time by Biden. And
01:10:00
that created a big rift in our society that we're still working through. The other rift that was created over the
01:10:05
past year, we've had many debates here about ICE and what some people, including myself, believe was a brutal
01:10:10
way to deport people and a cruel way to deport people. Some people might disagree with my opinion, but it seems
01:10:15
like Trump stood down on that and calmed it down, I think, because it was so unpopular, at least according to the
01:10:22
metrics that we saw from Pew and other uh studies. What do you think the road the the the path forward is now? The ICE
01:10:29
um aggressiveness has stopped. The border has been shut. We seem to be in a sort of holding pattern here which I
01:10:38
think was very wise by Trump to do that. Put us like calm the situation down. I would have liked to seen him calm it
01:10:43
down a little earlier. What do you think the go forward should be? If we go look
01:10:46
at 27, 28, 29, do you think obviously we all want legal immigration, but what do
01:10:51
you think we should be aspiring to do as a country now that the issue has seemingly is not boiling over? What are
01:10:58
your what are your thoughts? Jamaal, have you given it any thought? >> I think that Western countries are
01:11:02
losing their cultural representation that makes them unique and I think that that's very bad.
01:11:09
When my parents went to Canada and then stayed, I think part of what they were doing, whether
01:11:17
they knew it or not, was making an explicit decision to be Canadian first, which who happened to be of Sri Lankan
01:11:23
origin. When I came to the United States, I was making an explicit decision to become an
01:11:28
American and then my Canadian and then my Sri Lankan ancestry was second and third respectively.
01:11:36
I think if you seed your immigration, you're seeding your culture. And I think that people have a responsibility
01:11:44
to keep the values of what made that country unique in the first place. And you don't need to look too far because
01:11:52
you can look across the pond and you can look at the UK and you can look at Germany and you can look at France and
01:12:00
you have to ask yourself is the cultural deviation from where they were. Has it been better or worse
01:12:09
for those countries? And I think in general it has been worse. So I think people should want to come here to be a
01:12:16
part of the American ideal, the American culture. And so I think you have to have
01:12:22
a rational immigration policy that has compassion but ultimately favors this idea that you can't come
01:12:29
here to build your own version of your previous country here. So assimilation is part and parcel of a functioning
01:12:36
immigration system. I strongly agree. Freebrook, how should we go forward as Americans? We know Europe has had an
01:12:42
issue with uh or the populace is there I think are in agreement that they may have gone too far in terms of uh
01:12:49
unmitigated massive amounts of immigration specifically amongst groups that maybe are explicitly not wanting to
01:12:57
assimilate. What's your thoughts on what America could should do going forward? Do you believe we should be
01:13:02
>> It's a Yeah, it's a great question and I think there's an easy answer going back
01:13:05
to my makers versus takers point. If the primary motivation for an individual to
01:13:10
come to this country is to be given benefits, to be given payments, to be given social services, to be given
01:13:17
support, I think that that person should be denied immigration. If the primary motivation for the individual to come to
01:13:24
the United States is to work to progress themselves, to become a maker, to create
01:13:29
things, to work in such a way that they can generate income, save capital, buy things, advance their family's position
01:13:37
because they are denied those rights due to the tyranny of the place that they're
01:13:42
coming from, then they should be granted immigration status to come to this country and contribute. Because at the
01:13:47
end of the day, what people may not like to hear, which is the truth, is that every single one of those people is
01:13:54
actually net positive in growing our economy. And everyone thinks it's a zero sum game with respect to jobs in the
01:14:00
United States. And we shouldn't let people into this country that create jobs. But the fundamental truth is that
01:14:04
anyone that is productive, meaning they create more value than they take in, grows the economy and more jobs get
01:14:11
created. And overall, we all benefit from having those individuals here in this country. Well, I would say
01:14:16
>> so I'm all about I'm I'm all about discerning immigration not on the lines of some cultural like definition but
01:14:24
really around that simple framing of are you going to get social support services
01:14:28
or are you going to work and if you're going >> are you a train or a gain very simple
01:14:32
way if you are going to work you are going to culturally assimilate because that is what the United States is based
01:14:38
on which is productivity work hard create things and have individual agency have the ability to progress self in
01:14:46
this world, in this society. And if you take advantage of that opportunity, you're going to love the American ideals
01:14:51
that this country was founded upon. And there are ways to do that. Freeberg, you
01:14:55
can look at people's desire to learn the language and understand the political system and culture here. If you said at
01:15:00
the border, speaking fluent English, understanding our systems here, you know, and doing some learning before you
01:15:07
come here, that would give you more points in a point-based system. Canada, Australia, New Zealand, most of the
01:15:13
great functioning democracies in the world have pointbased systems. And part of that point-based system could be
01:15:19
you're provisional here and you get a visa for a certain amount of time provided you don't commit crimes, you
01:15:25
don't tap the welfare system, an entitlement system, and you do create jobs. We'll take an unlimited number of
01:15:32
Elon Musk, David Freebergs, David Saxes, and Chimath Polyhetas in my America as a
01:15:37
seventh, eighth generation American. And speaking for the people who've been here
01:15:41
the longest, we'll take an unlimited number of the besties I just mentioned and we should recruit them actively. We
01:15:48
should be trying to increase the population of the United States with those people. Every single person sachs
01:15:53
we take that's a high performer like yourself from another country like South Africa, Sri Lanka, uh, or pick the
01:15:59
country. Every time we recruit one of those people, that gives us a gain. It also in a super hyper competitive world
01:16:05
is a loss for China, for Iran, for wherever we can recruit the best and brightest who want to assimilate Saxs.
01:16:12
I'll let you have the final input here. What when you are um JD Vance's chief of
01:16:19
staff in uh 3 years, what will your advice to JD be? And how will you frame this during the election cycle? How will
01:16:26
you win my vote as a moderate? That'd be awesome. >> Chief of Staff David Sax. >> No, that's that's not going to happen.
01:16:31
But 100% going to happen. I talked to J. >> Well, I can give you I mean I can give
01:16:35
you the actual polling data. So, first of all, I think it's broadly popular in America to seal the border and not allow
01:16:43
illegal immigration. J, you're even on board with that obviously. Now, the only people who really aren't are these DSA
01:16:49
types who want open borders. They want to abolish ICE. They want ma mass amnesty. They are lunatics. Now
01:16:54
communist politics >> if you look at deportations about a third of US adults say all immigrants in
01:17:01
the US who here illegally should be deported and then half say some should be. So 32% are in favor of deporting all
01:17:09
illegals and 51% are in favor of deporting some and almost everybody in that group says that anyone who's
01:17:17
committed a violent crime should be deported. And then 16% which is the DSA group says none should be deported. So
01:17:24
deportations are popular provided that the illegals have committed violent crimes and have done other things that
01:17:32
people think should be unacceptable. But you're right. Only 32% believe that every single one should be deported.
01:17:40
That's from Pew. Now, I guess I agree with Freeberg's point about the incentives. I mean, if
01:17:47
you're coming here to get on welfare programs, forget about it. I mean, we should only allow people in if they're
01:17:54
here to work and be successful and are not a net like, would you say, a gain, not a drain?
01:18:01
>> I agree with that. And yeah, >> and even in the days of Ellis Island, >> you can have it
01:18:05
>> where the US was very accepting of immigrants, onethird of immigrants who arrived at Ellis Island actually
01:18:12
returned home. 2% were not allowed entry due to health or criminal problems, but
01:18:17
a third actually went back. And that was because there was no social welfare. And
01:18:23
if you didn't make it in the United States, you would leave. And now anyone can walk across the border. They
01:18:29
download an app and they get $5,000 a month in welfare. And during the Biden administration, they were getting luxury
01:18:35
hotel rooms and buses and flights all over the country. It was insane. >> So I mean, we got to stop that. If you
01:18:42
do get in, there should be some sort of moratorum on any kind of social program for maybe while you have a visa.
01:18:48
>> Can I just come up with that? I mean, it seems like the simplest thing you could
01:18:51
ever do is just say you cannot tax the welfare system. You cannot tax social security benefits, which I don't think
01:18:56
you can right now. You just can't tax anything. You can't take anything from the system for the 5 years you're here.
01:19:01
Period. Full stop. The end. And if that's okay with you, then come. And if you produce jobs and you create x amount
01:19:08
of economic activity, let's say you create $10 million worth of economic activity. When you hit $10 million in
01:19:15
economic activity, you are uh pushed to the front of the line and we review your
01:19:19
case on a one-off basis. If you hit 5 million, you go next. And that would be actually like let's put some goalposts
01:19:25
there. Let's set some targets. Hey, you come to the country. You're an Iranian Persian amazing entrepreneur. You're a
01:19:32
developer. You're an AI developer from Ukraine. You want to come here with your team. Great. What amount of money can
01:19:38
you raise on your way in the door? You've got $6 million in venture funding. Great. You go to the front of
01:19:43
the line. Let's go. Let's take these great entrepreneurs and creators and builders. Let's sweep the table with
01:19:49
them now that we've calmed down this rhetoric around immigration and shut the border, which was a key piece of doing
01:19:56
that. Uh, all right, let's keep moving. >> The GSA has calmed it down. They still
01:20:00
want to go back. >> They're lunatics. These [ __ ] commies. Honestly, >> the Democrat party.
01:20:05
>> I mean, >> let's talk about California, shall we, Jal? >> I mean, yes. Yes. Hold on a second. I
01:20:09
mean, if they take over the Democratic party, then I guess I'm a Republican. Colorado a 15ter term congresswoman 30
01:20:17
years list lost to a DSA so it's not just New York it's not just because of Zoran Mdani the DSA is very popular now
01:20:25
>> socialism is going to win any city >> this is this is a quick way to I mean to
01:20:30
Roana and Rahm Emanuel and to well I mean Rohan is a lost cause but to Rahm Emanuel to Shapiro
01:20:40
to anybody in Democrat is a lost cause. I've been telling you for a long time that you may
01:20:45
disagree with what Roana is doing and saying, but it's popular in the Democratic party. I think he's got his
01:20:50
finger on the pulse. >> Well, he wants to win. I get it. But I think you have to the quick win, the
01:20:57
quick hit, the quick bump of the fentinol or methamphetamine of communism and socialism to get you a
01:21:05
couple ratings point is not worth it long term. You will not win. You're going to lose every moderate, every
01:21:10
swing voter, every adult in the room is not voting for a [ __ ] socialist communist lunacy.
01:21:16
>> We'll see. >> They're going to implode the Democratic party. All right, we we'll leave off the
01:21:21
other um Scotas hottest summer for another time. >> Let's talk about Gavin with the good
01:21:28
hair Newsome. He says he just balanced the California budget. Nuome announced $351 billion in balance
01:21:37
zero deficit budget through 2028. Over to you, Freedberg, from your bunker, the last California stand in the Freedberg.
01:21:47
>> I'm going to call it I'm going to call this this the state of California. >> The state the state of California. The
01:21:53
lower >> the state of California. >> That's right. The state of California. >> The state of the state.
01:21:58
>> And there are five the state of the state. And there are five >> Got it. chapters to this ballooning
01:22:03
costs, the accounting tricks, the revenue dependency problem, and now the big exodus and spiraling that's
01:22:10
happening, and the looming unaccounted liabilities. These are the problems that California is facing as Gavin puts his
01:22:15
bow on his final budget, calls it balanced, and walks away while the House burns. California's state budget
01:22:22
ballooned from 215 billion a year in 2019, 6 years ago, to 355 billion today. So, a 65% jump in the state budget. And
01:22:32
there were a number of these kind of accounting tricks that took place to try and make it look like it's a balanced
01:22:38
budget this year. But there's revenue and there's spending. When they call it a balanced budget,
01:22:43
that's not the way we all talk about it in the business world where you have like a P&L, you have revenue and
01:22:47
expenses. They have revenue and expenses and the expenses exceed the revenue. And
01:22:53
that little difference they do a bunch of borrowing against and then they then they call it a balanced budget. So all
01:23:00
that they've done >> so they were able to raise enough debt to cover their whole question.
01:23:04
>> So there's somewhere between 20 and 40 billion of debt that they have a legal
01:23:10
ability to just pencil away and that is now a liability that the state has that the taxpayers are all going to have to
01:23:15
pay at some point that allows them to quote balance the budget this year. And I'm not going to get into all the
01:23:20
details of what went into this budget. all the money that's being spent. It's crazy. But I think that's what's worth
01:23:26
taking note of is the revenue dependency the state has. As you all know, the top
01:23:30
tax rate in California is 14.4% today. Personal income tax is $142 billion of the state's revenue. Okay, that is 211
01:23:42
billion of revenue the state generates. 142 of it comes from personal income tax. The top 1% 150,000 people pay 70
01:23:50
billion pay half of that tax of that income. So 70 billion of the state's 210 billion income comes from just the top
01:23:58
1% of payers, 150,000 people who are already paying the highest tax rate in the country. The top thousand people in
01:24:05
the state of California pay roughly 22 billion a year. So more than you know 11% of the state's income. The corporate
01:24:12
tax is one of the highest in the country at 8.9% $43 billion. So it's cheaper to
01:24:16
move out of state. If you look at some of the other states, Florida 5.5%, Texas, Tennessee is 6 and a half and
01:24:23
Texas is zero. So now if you kind of look at what's happening, there's a massive exodus
01:24:29
underway. the corporate exodus because of the high tax rates. Since 2019, at least 15 Fortune 500 companies have
01:24:37
moved out of California, moved their headquarters out. Not to mention all the local franchises and everything that
01:24:41
moved out. 2100 companies that are mid or large size have moved out of California since 2019. At least 5% of
01:24:49
California jobs have been lost as these companies have moved out of the state. And this is the more important and crazy
01:24:55
statistic. No one pays attention to this, but we are seeing an average annual exodus of 1 to one and a half% of
01:25:02
personal income, meaning the AGI, the adjusted gross income. So people that earn money, about 1 to 1.5% of that
01:25:09
income is leaving the state every year right now. That might not sound like a lot, but after 10 years, you've seen 15%
01:25:16
of the state's income leave. And obviously with the new billionaire tax that's being proposed, we're going to
01:25:20
see an acceleration as the numbers come out for 2025. There's, as we all know, personal friends that have left the
01:25:25
state in 25 and many more that'll leave in 26, including our friend David Sachs.
01:25:30
So there is a spiraling of a loss of revenue while there is a ballooning in costs. And this is fundamental to the
01:25:37
spiraling problem that's occurring. So now in order to try and make more money, the state has to go further down. They
01:25:43
can't just go to the corporates and the high net worth earners. They're now going to the average Californian. So,
01:25:47
they're increasing sales tax rates this year by creating a new sales tax on software. So, if you buy Microsoft Word
01:25:54
or you pay for your Gmail subscription or even your Chat GPT account, you have to pay an 8% sales tax on that this
01:26:00
year. They're estimating that that sales tax is a brand new one. It'll generate about a billion dollars a year in
01:26:06
incremental revenue for the state. They're also creating a new tax on healthcare insurance that corporates are
01:26:11
going to have to pay that's ultimately going to pass down to the individual. That's going to create another $2
01:26:15
billion a year in taxes on health insurance. The final thing that they did, which we all know about, is they
01:26:20
made the temporary top income tax bracket permanent. So, you know, it was only meant to be a temporary to bridge a
01:26:27
major budget deficit problem back in 2012, 13.3%. And that became permanent and now it's increased to 14.4%.
01:26:36
So, this is the spiraling that takes place. The costs are ballooning, the revenue is leaving, and now there's a
01:26:42
scrambling underway in the state of California to try and make up this difference. And they're still falling
01:26:46
short. In the out years, 2028 29, the projection is $40 billion a year in budget deficits every year. Now, looming
01:26:54
over the state of California is the fact that there's this insane unaccounted debt problem. California already has
01:27:01
$1.4 trillion in public debt. The state has 500 billion and the local governments have another, call it 800
01:27:07
billion. The state has a reported $664 billion in unfunded pension liabilities. That number by many estimates is closer
01:27:16
to 1.5 trillion. And then there's a retirey healthcare obligation deficit of $175 billion. You
01:27:24
add this all up, there's somewhere around 1.5 to2 trillion of incremental liabilities that the
01:27:31
state has to come up with. And remember, all of those pension liabilities, that trillion dollar plus of pension
01:27:36
liabilities sits senior to the bonds of the state of California because of something that's known as the California
01:27:42
rule. So the state has this looming kind of cliff ahead of it and there's already
01:27:48
this spiraling problem where they're increasing costs because of the demands of the unions and the cost of living and
01:27:53
so on and there's an exodus of corporations and high netw worth individuals that pay the taxes. So, I
01:28:01
would argue that Governor Nuomo did not exactly put a bow tie on the budget and create a balanced budget and leave
01:28:07
California in a great state as he is wrapping up his final term and his final budget here in the state. But in fact, I
01:28:13
think that this state, which, as I've said in the past, threatens to bring down the union, is on the brink of
01:28:19
defaults that are going to be so significant that if the federal government was called in to bail them
01:28:25
out because a state cannot declare bankruptcy legally, that all of the red states and all of the other taxpayers in
01:28:30
the United States will say, "Why the hell should I pay federal taxes to bail out California?"
01:28:35
And I think that's what the story will be over the next decade with the state the fiscal condition
01:28:40
>> AOC and Mandami president vice president AOC Mandami can't be VP he wasn't born
01:28:44
here >> I do predict yeah so I predict AOC will be president I mean obviously there's a
01:28:48
lot of dispersion in what could happen here but I would say AOC would be my front runner based on the the movement
01:28:53
for the extraordinary cost of living in the United States right now largely because of government spending
01:28:59
>> the fact that people are seeing this wealth disparity those two things are going to continue to drive socialist
01:29:04
movement over the next 24 months. And if AOC comes in and they bail out California and they federalize
01:29:10
California's liabilities, you're going to see parts of this union, the red states as they're called, the Texas of
01:29:17
the world, that are going to say, why should we be part of this union anymore? >> We don't want to pay for all of these
01:29:22
liabilities that these blue states have accured and that we are now being asked to bail out. And that is where I think
01:29:29
you face a crisis of the union in the years ahead. >> Okay. Sachs Texas the first Yeah. Yeah.
01:29:39
>> In this scenario, I mean, it's pretty farfetched, but it's not farfetched that
01:29:44
you would you would need a federal bailout to get California out of this in the next couple of years, and the
01:29:50
government needs a bailout itself with the massive uh we're going to add two to three trillion dollars a year for the
01:29:56
next couple years. So, go ahead, Shimothia. >> There's something that's going to
01:30:00
happen. I think that Freeberg is right up until the point of the federal absorption. I think what'll happen is
01:30:07
that you will wipe out the California pensions and you'll wipe out the pension obligations and you'll do some sort of
01:30:13
negotiated settlement. And I think that's where you're going to have a wholesale replacement of the California
01:30:20
Constitution. You'll have a complete redistricting. I think you will have a complete red wave
01:30:26
in the state. and you will disassemble and dismantle everything that caused California to be
01:30:34
so poorly run that got them to this place. So I I agree with David. All of the issues that he painted are it's
01:30:41
unbelievable. By the way, there's an incredible tweet. I think her name is Joanna Ives. Maybe Nick, you can find
01:30:46
it, but she compared the last eight years of Dantis versus JB Pritsker. And I asked my team, hey, can you create a
01:30:54
table that actually adds Newsome to the list so that you can do the three-person
01:30:57
comparison because it's quite staggering how broken these blue states have become
01:31:04
and how they continue to try to spend their way out of what is being shown is just general incompetence. They're not
01:31:12
good at managing. They're not good at running things. They're not good at delivering results. And unfortunately,
01:31:18
until they feel the brunt of what mismanagement actually means, which is that all the people that have paid into
01:31:26
this system, sadly and unfortunately and avoidably will be the ones that get punished, we will unfortunately not
01:31:34
learn in California. Find this tweet and read it and you would just be shocked at
01:31:39
how unbelievably good Dantis has been in running Florida and how unbelievably tragically bad JB Pritsker is. And then
01:31:46
just ask your favorite AI, compare California on the spectrum of Florida versus Illinois. And what you'll find is
01:31:56
California is even worse than Illinois in most of those comparisons. >> Sax, you want to uh give us the last
01:32:03
word here? >> Yeah. I mean, I guess my decision to go to Texas is looking better and better.
01:32:09
Yes. >> I mean, I said when this when this BTA thing, the so-called billionaire tax act
01:32:14
was proposed >> and everyone was saying that's very low probability this happens. I was like, I
01:32:19
don't really see why it would be low probability. I think it's going to get on the ballot. All you have to do is
01:32:23
collect a certain number of signatures. It's not that hard. And then once it's on the ballot, I don't see why it
01:32:28
wouldn't pass. And everyone's like, "No, no, no. Gavin Newsome has assured us that he'll kill it. He'll make some
01:32:34
backroom deal, get rid of it." And on the day, the final day where BTA was going to be dropped from the ballot or
01:32:43
not, Newsome came out with a video on X where he basically endorsed more taxes on billionaires. And this was right
01:32:50
after the the DSA candidates won in New York. So, he sees the writing on the wall. He's not going to come out. By the
01:32:57
way, the the problem with the billionaire tax is that if it does get passed and it isn't obiated by the three
01:33:04
other ballot initiatives that actually negate it, you're going to have two or 300
01:33:11
individual lawsuits that will take a decade to meander through the courts and ultimately I think a lot of people have
01:33:17
standing here. And I just wonder what will they have proven at that point because it'll be a decade later. It'll
01:33:24
be 2036 and the California government will have had to spend billions of dollars on
01:33:29
legal fees to to fight all these billionaires all all to end up losing at the Supreme Court. Anyways,
01:33:36
>> by then they'll probably have like a federal wealth tax or something like that.
01:33:40
>> The tragedy is that California at various points in recent history has had a surplus because of the massive tax
01:33:47
base from Silicon Valley. They could have managed this. Their incompetence, their selfishness, and their corruption
01:33:55
is what has caused this. They could have, if any state could have balanced its budget, it is California because of
01:34:01
our massive tax receipts. >> There needs to be a hard landing. I really hope it doesn't result in the
01:34:08
wipeout of these pensions. I think the pensioners should be extremely upset >> with California Democratic politics.
01:34:15
Look, this idea that somehow things get bad enough where you get a red wave in super blue California and everything
01:34:23
gets magically fixed, not going to happen. What's going to happen is a blue state's going to get bluer and
01:34:28
>> that's right, socialism. >> The remedy for more and more socialism is going to be to intensify the
01:34:33
socialism and there's going to be a giant final confiscation. >> Then we're going to deal with a
01:34:38
California bankruptcy in the next 10 years. And it is unfortunately >> No. and it will unfortunately result in
01:34:45
a complete restructuring of those debt and obligations. And this is where I think you're going to wipe out a lot of
01:34:51
these things that people are guaranteeing. I don't think you're going to see these pensions honored. If you
01:34:55
just look at the budget per person, uh places like Texas and Florida spend about $5,000 per uh citizen on services.
01:35:04
You know, if you divide the budget by the population, and then you look at New York City, it's close to $13,000. And
01:35:09
you look at California, close to 9,000. And you simply do not get two or two and
01:35:14
a half times or 2.x times the value living in those states in terms of services.
01:35:19
>> The problems are all worse. >> And it's competence. Yes, it's worse. And it really the more you look to that
01:35:24
state government to solve your problems, the more problems you will have in Texas
01:35:30
if you got a just seems these blue states attract these derelict incompetent people. I
01:35:38
mean, unbelievable. Incompetence. They're so incompetent. My god. This has to do with competence. And it also is
01:35:46
the populace being self-reliant. If you live in Texas, >> when you get your ranch, they say, hey,
01:35:53
>> you know, here's your ranch. And then it's up to you to figure out, hey, who's
01:35:56
going to pick up my garbage? And oh, do I want a well? Do I want to use a city water? What's my energy situation going
01:36:01
to be? And people make these kind of thoughtful decisions with us. They don't look to the state for every single
01:36:07
handout or handholding. There's a ruggedness and a rugged individualism and self-reliance that makes the state
01:36:16
not have to tax everybody to 54 60 65%. I guarantee you Freedber and Shimoth you
01:36:23
will be paying 60 to 65% tax if you stay in California. And I tell you New York it's at 54 right now. I guarantee you it
01:36:31
breaks 60 by the or it will be proposed to break 50. We'll see if they stop it by the end of Mandami's reign in my
01:36:38
hometown. >> Look, let me let me tell you what leftism is, okay? It's basically just
01:36:43
people trying to string together words to justify taking your money, okay? And they'll they're just experimenting. It's
01:36:50
like monkeys typing on a keyboard with words to try and hack everyone's brains to basically figure out how to steal all
01:36:58
of their money. And some of those words are virtue signaling. Some of them are guilt, tripping, and moral accusations.
01:37:05
>> Whatever it is, there's always some new words. >> But it's never to solve a problem. In
01:37:10
fact, all the problems just get worse. >> It's just to figure out how to separate
01:37:14
you from your money. And I had this >> thought when I saw that like McKenzie Bezos, you know, just giving away like
01:37:22
>> 26 billion uh 26 billion of Jeff Bezos's money to all these leftwing causes. Hey,
01:37:28
she earned it too. >> And well, but putting that aside, she gave away all this money. I'm like, "Oh,
01:37:34
okay. So now, like world hunger is fixed, right?" Because remember how they told us that it was because of Elon's
01:37:39
Musk re that he refused to hand over, he didn't fork over a big chunk of his money. And with that money, we could
01:37:45
have fed the whole world and solved world hunger. So why isn't the problem solved now that she just donated 26
01:37:51
billion? That's what they said Elon needed to do to solve world hunger. They always treat your money as if it belongs
01:37:56
to them or it belongs to the government and it's your greed and selfishness that's creating all these problems. None
01:38:01
of these problems have been fixed. What's the proof that any of that money did anything? All it did was go to line
01:38:08
the pockets of all these leftist NOS's and you know buy them more houses or or pay them
01:38:15
>> better off investing in oh no problems companies if if literally I I hate to break this news. What does the poly
01:38:22
market say on Gavin's chances of uh >> Oh, he's down to becoming He's plummeting. He's plummeting.
01:38:29
>> Yeah, Gavin's down to 20. He peaked at 40. >> That's quite high, though. >> I I I mean, I think people are real
01:38:35
realizing it's a great haircut and uh you know, a really smooth talker, but just like he stole other things in his
01:38:42
life, he's going to steal your money. Period. Full stop. I'm not going to get into detail. I think he's lowering these
01:38:48
polls because the Democratic party is moving left is moving to the TSA and people think he's too moderate
01:38:57
>> is going up and this John Oaf guys come out of nowhere. They're all to the left
01:39:00
of Gavin Newsome's problem and this is why he put out that video on the day that he was supposed to
01:39:06
kill the BTA. He puts out a a video endorsing billionaire taxes is because he needs to position himself to being
01:39:14
more DSA adjacent. Well, to be fair to Newsome, he also on the same day said he's against the billionaire tax act and
01:39:19
he's voting against it. This was his hedging video, which is to to kind of zoom out his policy on, hey, this BTA
01:39:26
doesn't make sense. But I do agree we need more taxation, I think, was kind of the intent. Also, he failed he failed to
01:39:32
kill BTA. Everyone thought all these California billionaires and these chat groups thought Nome would be their
01:39:38
savior. >> That's right. >> And that was a big mistake. >> What's it like, Sax? What's it like in
01:39:43
that group chat? Tell us about that group chat. >> What's it like in there? What's the
01:39:47
meme? What's the meme like in there in the billionaire group chat? Go ahead. >> I It's It's 8:06 p.m. here. I have a
01:39:56
meeting I have to go to. I want to say I love you and I miss all of you. >> And I want to say
01:40:01
>> a huge happy birthday to the most incredible country in the world. >> I could not have imagined my life being
01:40:11
any better in any place. I feel so blessed and so lucky to be an American. >> I feel so thankful. I feel so grateful.
01:40:18
It really is the best country in the world. >> We have to fight for it. >> Yeah. We have to fight for it.
01:40:22
>> Fight for it. >> Yeah. And on behalf of those of us >> Happy birthday, America.
01:40:26
>> Happy birthday. And for those of us who have been here for seven generations,
01:40:30
>> we think you're doing a good job. We're going to let you stay here. We're going
01:40:32
to let you stay on behalf. >> How can we take away your citizenship? That's what I want is a law where we can
01:40:38
revoke JL citizenship. >> Citizenship, [ __ ] I've been here >> which the four of us should be deported.
01:40:44
I think you would lose. >> Oh no. >> We'll find some constitutional loophole. >> Enemy Aliens Act.
01:40:51
>> Bye, Jason. >> All right, feel free. On that note, happy birthday, America. And uh 250 more
01:41:00
great ones. Best country in the world. Give a [ __ ] you think democratic socialist commie scumbags. All right,
01:41:06
we'll see you next time. Bye-bye. Love you. Bye-bye. Let your winners ride. >> Rainman David
01:41:19
and it said we open sourced it to the fans and they've just gone crazy with it. Love you queen of
01:41:32
>> besties are my dog taking notice your driveways. Oh man, my habitasher will meet me up.
01:41:42
>> We should all just get a room and just have one big huge orgy cuz they're all
01:41:45
just It's like this like sexual tension that we just need to release somehow. >> Your feet.
01:41:55
We need to get mercy. >> I'm going all in.