00:00:00
All right, everybody. Welcome back. Welcome back to the number one podcast in the world, the All-In podcast,
00:00:05
episode 278. And uh Freeberg, he took a mental health day today after the socialist sweep in
00:00:13
New York City. So, we invited two guests. They both said yes. Travis Kalanic is here from Adams. How you
00:00:19
doing, brother? >> I'm good. I'm good. Good to see you. >> Good to see you. Good to see you. And
00:00:25
after a triumphant week at Starbase, the one, the only, everybody's favorite, Gavin Baker of a treaties management.
00:00:34
How are you doing, Gavin? >> Great, man. Thanks for having me. >> You're still floating on Cloud9 after
00:00:39
the SpaceX IPO. Yeah, >> that was a very special moment and it was a culmination of, you know, some
00:00:45
decades of hard work, >> but man, to quote uh Bill Bich onto Cincinnati. >> Yeah. Yeah. Exactly right.
00:00:54
>> Yeah. Zero zero is uh the how the Knicks talked about the next game in every series. They're like, "It's 00. We come
00:01:01
into this as if it's like the first game of the series, even if we're up two games or three games. The socialists
00:01:06
have swept New York in the congressional Democratic primaries." On Tuesday, New York City Mayor Mandami went three for
00:01:14
three in the candidates, which he endorsed, and they all won their primaries. 10th district leader Brad
00:01:22
Lander defeated two-term incumbent Dan Goldman. 10th is one of New York's richest districts includes the West
00:01:27
Village. All those townhouses, uh, Wall Street, Dumbo, Cobble Hill, Carol Gardens, Park Slope. That's some weird
00:01:34
geography put together there. In New York's 13th, Jiovalier beat a fiveterm incumbent who was backed by House
00:01:43
Speaker Hakee Jeff. And apparently the socialists are coming for him. 13 is one of the poorest districts. Harlem and the
00:01:50
West Bronx, the Boogie Down Bronx. She's a 32year-old Democrat socialist with a history of spicy remarks.
00:01:59
New York's seventh district. Claire Valdez won the open seat over the handpicked successor. The incumbent
00:02:06
seven is a DSA stronghold. Bushwick, Williamsburg, Long Island City, Greenpoint, known as the Kami Corridor.
00:02:14
A lot of hipsters and baristas with suspenders in that neighborhood. According to our partners poly market,
00:02:20
this was a pretty big upset. The Mandami sweep chances were just 26 before election day. Yeah, that would be like
00:02:28
the trifecta there if you were gambling. These candidates, just like Mandami did,
00:02:33
a lot better with younger, college educated, and highincome folks, folks who can afford to be socialists.
00:02:39
And these are all safe Democratic seats. The DSA will very likely win. So the DSA
00:02:43
caucus, >> that's the best way you just said, people who can afford to be socialist.
00:02:46
It's always It's always the the rich poors, you know, they're rich but they pretend to
00:02:51
be poor. It's perfect. >> I mean, this is the history of our country from the 30s, the 50s, Red
00:02:56
Scare, Black Panthers. I mean, we'll get into it, Shimoth, but basically, if you're intelligent and you get into
00:03:02
business, you become a capitalist. If you're super intelligent and you go into academia eventually, you have the luxury
00:03:08
of belief that socialism is awesome. What's your take here, Chimath? And then I'll go to you, Sax Poo, cuz I know I
00:03:13
can see you're chomping on something over there. you're just ready to go. >> Honestly, I think that we are losing the
00:03:19
script and part of it is because we've been our own worst enemy. I'll just keep saying this that I think
00:03:27
AI is a very good prism into this problem. I think AI is the greatest economic leveler
00:03:35
we'll ever find in our lifetime. I think it's the thing that can create the greatest amount of equality. I think
00:03:41
that it can even the starting line for every single person on earth. But we've done such a poor job in representing it,
00:03:49
in bringing it to market, in talking about it. We've let all of our own personal trials and tribulations and
00:03:54
insecurities and fights spill out into the open. As a result, Silicon Valley has lost even more credibility with the
00:04:01
people at large. And in that vacuum, what other people can paint is a picture of how anything other than what
00:04:08
capitalism looks like today is a better version of what they see. And this is why you're seeing, I think, a lot of
00:04:16
these people get a lot of momentum. I think if you look at some of the key congressional races, they were a
00:04:21
referendum on AI. And the good news is we were able to hold the line in some of these key places, but just barely. In
00:04:29
Utah and in New York, there was a couple of very important races where it was essentially anthropicfunded
00:04:38
anti-AI groups, which is again insane against in some ways open AAI funded proA groups
00:04:46
and the pro-AI groups won. You just explained that it's a a great leveler. Can you give a couple of examples and
00:04:52
just expand on that briefly? How is it a great leveler? I mean I know but for the
00:04:56
audience's sake you've said this a couple of weeks now so I think unpack it a bit.
00:05:01
>> Um the best way to explain it is I think that the first real major unlock of economic productivity was when the
00:05:10
internet and specifically Google went and harvested and collected all the world's knowledge and all the world's
00:05:17
information and they made it available via search. What we figure out though 25 years later, despite the fact that they
00:05:23
built a great business, is what was missing was then being able to take that knowledge and information and transform
00:05:31
it into expertise and intelligence. And that's effectively what AI does. It takes that world's knowledge and it
00:05:38
allows you to act upon it so that every single man, woman, and child has an equivalent Travis Kalanick in, you know,
00:05:45
as his co-founder, a superfounder, this brilliant person that can think through all your problems, can outgineer people,
00:05:52
can outthink people, and they sit beside you and you have that. And there is no gatekeeping that can prevent you from
00:05:59
having that. And so now you're only as good as your ability to direct that energy into something productive that
00:06:06
you value. That is an incredibly powerful thing. And instead we've gotten caught up in dumerism and jobs being
00:06:13
lost and you know water being consumed. All of which are lies. All of which are complete fabrications and
00:06:20
misinformation. And these have been created in order to specifically help one small set of
00:06:28
actors inside the AI race. and it's been fed and funded by those folks. So in in
00:06:34
this vacuum, Jason, we've allowed all these other people to paint the other version. And right now, the other
00:06:40
version looks way more compelling than the current version because the current version has very poor brand ambassadors.
00:06:47
Sax, what's your take on this? I guess some people are looking at this as the left's
00:06:54
version of the populist takeover that Trump did over the last 10 plus years. What's your take on what
00:07:00
we're seeing here with socialism and its amazing appeal and it's winning at the uh election box?
00:07:09
>> I think there is some truth to that. I mean, I think the choices of the future
00:07:12
are going to be communism or if you want to call it socialism in the Democrat party or nationalism in the Republican
00:07:18
party. I mean, that is where we're headed. Those are the two populist directions. But let's look at what these
00:07:24
DSA candidates stand for. So, let's look at what their platform is. They actually
00:07:32
say they want to abolish the Senate. They want to abolish the carceral state. That means basically police forces and
00:07:40
prisons. They want to abolish ICE and grant amnesty for all. They do not support any deportations whatsoever.
00:07:48
They want to replace the president and supreme court with an executive and judiciary that is chosen by and
00:07:54
subordinate to Congress, which basically now I guess just means this house. And with respect to House elections, they
00:08:01
want to abolish the electoral college. They want to replace the two-party system with a multi-party democracy. And
00:08:07
they want to expand the House of Representatives, implement proportionate representation and rank choice voting in
00:08:12
all elections. So this is would be a total makeover of our constitutional system. They want to free Palestine.
00:08:19
They want public ownership of major corporations. They want to defund the Department of War. This is a very
00:08:25
radical organization. And you would laugh at a lot of these types of proposals, but you can't really laugh at
00:08:32
it anymore because these guys are taking over the Democratic party. And you can see the Democratic establishment is in
00:08:39
complete panic right now because they have lost control of the party uh to Zoram Mdani and his his allies. So
00:08:47
Jason, like you said, I mean, let's take this one race here, New York 13. You've
00:08:52
got this ally of Hakeem Jeff, longtime incumbent congressman, Espalot, I guess is is his name. He is
00:09:00
the chair of the Congressional Hispanic Caucus. And he was defeated by an unemployed 32-year-old PhD candidate.
00:09:07
She's never had a job. She's been in college for 10 years, I guess, writing this PhD thesis.
00:09:14
>> And I think even by DSA standards, she might be kind of a lunatic. So, she has
00:09:19
declared that she wants to end Western civilization. She wants to eradicate Western civilization.
00:09:24
>> Wait a second. What? >> We're soaking in it. >> Yeah. She actually said she used the
00:09:30
American flag as a napkin to clean her hands. >> Oh, >> she attended a rally one day after
00:09:38
October 7th celebrating the slaughter of Israeli civilians. I mean, she's very um
00:09:42
pro Palestine, but even to the point of celebrating Israeli civilian deaths. She
00:09:48
calls white women ugly colonizers. She's called for the complete defunding the police and abolishing all prisons and
00:09:54
borders. Doesn't want a single deportation. hates the police, openly calls them pigs, or has on social media
00:10:00
before, calls US service members war criminals, and says the US is a disgrace of a country. She's written favorably
00:10:08
about communism and seizing the means of production, and on and on and on. So, this is basically the new Democratic
00:10:16
party. It's going to be even if the Democrats do take the house in November and Hakee Jeff becomes speaker, this is
00:10:22
going to prove to be a huge headache for him managing all these new DSA members >> because they do not actually see the
00:10:29
traditional establishment wing of the Democratic party as an ally. They see it as an obstacle. This is the DSA co-chair
00:10:36
Josh Block said, "We're using the Democratic Party as a ballot access vehicle." >> Oh my gosh.
00:10:43
>> Not because we share its goals. We build our own organization, >> get elected under the Democratic label,
00:10:49
caucus with Democrats when it's useful, and push our own agenda from the inside.
00:10:53
We see the Democratic establishment as an obstacle, not a home. So, the DSA is coming from the Democratic Party. It
00:11:00
controls the base now. It's where all the energy is. I think this takeover will continue. I think the DSA will
00:11:06
gradually take over more and more of the Democratic party. And all I can say to these establishment Democrats is play
00:11:13
stupid games. win stupid prizes. You supported this open border policy that brought in this wave of mass migration.
00:11:21
That is a huge part of the base of this new DSA wing. Ma'am Donnie would not have won the mayoral election in New
00:11:28
York if it had just been nativeorn New Yorkers voting. It was the mass migrant vote in New York that swung it to Ma
00:11:34
Donnie. It's not exclusively the DSA base. It's the migrants plus these overeducated
00:11:42
white progressives who I say overeducated because they're more downwardly mobile. They end up going to
00:11:50
work in academia or NOS's that kind of thing. They're hard left-wing. >> They're kind of the vanguard. And it's
00:11:57
this combination of these, you know, recent college graduates who are kind of the organizers and this migrant movement
00:12:05
who are really taking over the Democratic party. But again, this all goes back to Democratic Party policies.
00:12:10
They may not have intended for the staff and they may not have intended for them
00:12:13
to lose control, but it was their open border policy. It was also the fact that they have cracked the melting pot and
00:12:20
the policy we had for many years in America of assimilating migrants, right? Immigrants.
00:12:26
>> Yep. >> That all got cracked by multiculturalism and wokeism. We don't really do that
00:12:30
anymore. So now you've got these candidates like Shioalier openly declaring they're not just
00:12:36
anti-American, they're postamerican. They don't have any respect for the American system, our constitutional
00:12:42
order, our free enterprise system. They want to introduce something in much different and it's going to look a lot
00:12:49
like the countries where these migrants are coming from. >> Yeah. So again, you know, if you import
00:12:54
massive numbers of migrants, don't assimilate them into our system, and then you have these Marxist leaders,
00:13:01
you're going to end up with the American system coming to resemble the countries
00:13:05
from which >> it's really interesting. You mentioned like how she wants to get rid of all of
00:13:09
these, you know, delete all of these uh units. It's very similar to Trump's, you
00:13:14
know, he wanted to delete a separate set of things like the IRS and USID, Department of Education, EPA. They each
00:13:21
use that playbook as well, like, "Hey, we're going to delete some of this stuff." Travis, you uh famously had uh
00:13:27
the fountain head. >> Trump didn't want to delete the US Constitution. >> No, definitely not.
00:13:33
>> Travis, you're used to Slight difference. >> Slight difference. I agree. He didn't
00:13:36
want to delete Western culture. >> I don't think he wants to delete Western culture.
00:13:40
>> No. Uh >> you might want to delete USAD because it was a festering. It was basically a
00:13:46
front for all these >> NCOs. NCIA. Yeah. By the way, you may you may have noticed that in South
00:13:52
America, we've had seven elections where the countries have swung to the right. And I think a big part of the reason why
00:13:58
might be you don't have the USAD there conducting all these uh left-wing regime change operations.
00:14:06
>> Pola's polared disappeared and the incentives disappeared with it. >> Travis, you uh famously had The Fountain
00:14:14
Head and Ran's novel as your Twitter avatar for a decade or two. Uh it was no they've been writing about it for a
00:14:23
decade or two. This is like a six-month thing where I would read a book and then
00:14:27
rotate at my avatar. >> Yeah. >> You had Hamilton for a while. I had Ender Game for a while. But yeah, it
00:14:34
became >> I mean you having known you for a while I think believe in the individual and
00:14:41
their exceptionalism and and trying to have a little ruggedness there. This is obviously the the exact opposite. What's
00:14:48
your take on what's happening in these pockets? Cuz it's not national yet, but it's definitely notable.
00:14:56
>> All right. I've um I've got I've got sort of two apherisms for us. Okay. First is
00:15:05
truth and justice is the immune system for society. When the immune system is suppressed,
00:15:11
all the social ills flare up. Okay? So if you're seeing us losing truth like social media, mainstream
00:15:20
media, whatever you call it, like that's an early indicator or bad things happening in society.
00:15:29
Okay? And it's not just like social media, mainstream media, it's just everything around us. Um, and the same
00:15:35
on the justice side, too. If people commit crimes and there's no consequences, it's a nice early indicator. So you can
00:15:42
kind of watch you can watch these things and say is truth winning today or is it losing?
00:15:49
>> Is justice winning today or is it losing? And what is the trajectory will tell you? Are we going to get worse? Are
00:15:56
we going to get better? >> Right. Co makes and the Fouchy case now we're seeing a lot more information come
00:16:02
out. We have a really hard time getting the truth and maybe we'll get justice eventually. It seems like we're getting
00:16:08
a little more truth 5 years later, but it's like a perfect example of what you're discussing. Yeah,
00:16:12
>> there's a lot packed into that, but you can unpack it. If we had a lot of time,
00:16:15
we could unpack it. But those become the sort of atomic things that you look at.
00:16:20
And then the other thing I'd say, and and some people get a little surprised when I say this, which is
00:16:26
communism is in is in all of us. Communism is in is in our blood as humans. And people go, "What the hell
00:16:34
are you talking about? You're crazy. What do you mean?" Well, I'm like,"Well, have you have you ever in your life been
00:16:41
lazy?" And everybody's like, "Yes, I've been lazy in my life before." I said, "Have
00:16:47
you ever in your life wanted something for nothing?" Yeah. The difference is, do you make
00:16:55
that a way of life? And when you have ecosystems that essentially allow you to do both of
00:17:03
those things without consequence, those ecosystems get a critical mass and start
00:17:08
taking hold. And I think that's what we're seeing. >> Well said. Yeah, I think it's well said.
00:17:11
Uh Gavin, I I know you uh I don't know you to talk too much about politics, but talk about markets, but these two things
00:17:18
do relate, I think. So give us the economic perspective here on why this is happening because obviously different
00:17:27
generations have had different economic experiences. So I don't want to lead the
00:17:30
witness here but I have had conversations with you about this before. >> Yeah. Well so I do think um
00:17:37
I obviously AI is I think going to be the defining political issue of the midterms and for sure the the next
00:17:44
presidential election. But I feel like what is going on with the DSA is really a fusion of two
00:17:50
things. So, the Democratic Party, I I I was a Democrat for most of my life, and it was the party of the working-class
00:17:57
people that was trying to create opportunities, you know, maybe level out um equality,
00:18:06
you know, help black Americans, help Hispanic Americans. And, you know, you can agree or disagree
00:18:14
with their methods, but I think those are all noble goals. And to a large extent,
00:18:22
none of those are really present in the DSA. None of them. If you look at who voted
00:18:28
for which candidate, the pe the voting base of the DSA are relatively wealthy white liberals
00:18:38
who are downwardly mobile. They're losing votes with working-class people. They're losing votes with poor people.
00:18:45
They're losing votes with black Americans. They're losing votes with Hispanic Americans
00:18:50
and those, you know, the the people that the Democratic Party, I think, for a long time tried and whether they failed
00:18:59
or succeeded is up to tried to represent and give a voice to, they are not of interest to the DSA.
00:19:06
And I think the DSA is is dangerous. It's tragic that we are electing profoundly anti-American
00:19:15
candidates who in some cases have called for violence uh to eradicate America. And I just think there's there's a whole
00:19:24
class of people who went to an elite school. They grew up in really nice circumstances
00:19:32
and uh instead of going into industry, they went into this kind of giant NGO nonprofit machine
00:19:43
and their outcomes have been very different from people who did productive things for the world. You know, I've
00:19:50
said on this show many times, Elon has done more to decarbonize the planet than every activist combined.
00:19:58
>> Yeah. times 10 by the way >> times 10. And I think these NOS's and the fact that the government is
00:20:05
increasingly outsourcing a lot of funding to them is a really big part of this problem and
00:20:15
they are pursuing policies. You know, I think we maybe talked about it last time, but the curly effect, a mayor of
00:20:20
Boston, you pursue policies that you know are going to be disastrous for this your constituents, but they drive out
00:20:29
your rivals. And then you can give jobs, $600,000 a year jobs, run an NGO that does nothing productive to your friends
00:20:40
and allies. And it's really organized corruption happening at a massive scale. >> You know, the one thing, whatever,
00:20:49
whatever you want to say about the Trump administration, a lot of people are up in arms about this, you know, new plane
00:20:53
he's getting from the Qataris. >> Well, well, one, sounds like we needed a new plane, but two, it's right there for
00:21:00
you to see. I think there are tens of billions, maybe hundreds of billions of these payments flowing to these NOS's.
00:21:09
And if you look in California and New York, it's um like I think the per capita spending on
00:21:17
homelessness has more than doubled and it's all gone to NOS's and outcomes have gotten worse. In California, it's like
00:21:24
quadrupled and outcomes have gotten worse. So, I think this is this is dangerous
00:21:32
and I hope that the mainstream Democrats can find a compelling candidate because
00:21:39
I think the reason that the DSA is increasingly ascendant is not because of their ideas. I think their ideas appear
00:21:46
to this appeal to this very nar narrow subset of downwardly mobile rich white people who you know increasingly maybe
00:21:53
live in you know quasi gental poverty because they they're not doing productive things. They may be
00:21:58
well-intentioned but they're not doing productive things and can find a compelling candidate
00:22:04
because I think the reason they're ascendant is Zoram Mani. I think he is one of the most talented politicians
00:22:11
I've ever seen in my lifetime. You know, he can give a great speech. He's good in
00:22:16
an interview. He can tap into all of this. He's kind of a chameleon who can shift. But I think he is a singularly
00:22:25
talented politician and he is the reason that the DSA is ascendant. not their ideas or not dissatisfaction
00:22:33
with AI, but it's him and there's no one else in the Democratic party. I used to
00:22:41
think AOC was by far the most talented Democratic politician. >> That was the warm-up. Yeah,
00:22:45
>> she was talented. She's nothing like Zoron from a political >> Zuron is clearly talented. And if you
00:22:51
combine his charisma and ability to communicate with generations like we're now like have two of these lost
00:22:58
generations that feel like they're going to do worse than their parents. And if you look at housing and college debt,
00:23:06
healthcare, they don't believe that they can participate in the system. They feel
00:23:11
the system's rigged. And if somebody comes along who speaks to them and they have no conception of socialism and what
00:23:18
happened, you know, in Germany or during the Red Scare, they have no idea what socialism or communism is. They have
00:23:24
this incredible wrapper they've put around this, which is, you know, they're democratic socialists. It's like it's
00:23:29
it's the coke light. It's it's a coke zero of socialism. It's actually just communism. They want to literally seize
00:23:37
people's assets. They want to seize half of, you know, these compan stock. They want to seize their wealth. They they
00:23:42
just want to take from the people who have made stuff as we talked about last week and we've done nothing to change
00:23:49
their mind. We haven't made more houses. We haven't made college more affordable.
00:23:53
Healthcare is a disaster. Inflation's a disaster and starting pointless wars. Obviously, the support of Israel is
00:24:00
somewhere in here as part of say something. >> Yeah. Yeah. I was on Megan Kelly yesterday
00:24:07
and I had this theory that I've been kind of working through and I'll just share it with you guys because I'd love
00:24:12
your reaction. If you look at the scourge of socialism, if you take the rhetoric away and you
00:24:20
actually look at the outcomes, there are three countries that I think have veered
00:24:25
far towards socialism well before the United States. Canada, the UK, and Australia. And I think when you look at
00:24:34
any sort of reasonable measure of their progress, it's been an unmitigated disaster. So all of the virtue signaling
00:24:42
on social issues, all of the virtue signaling on immigration and open borders, all of the virtue signaling on
00:24:48
climate change has left each of those three countries in some state of disrepair with enormous amounts of
00:24:56
infighting, tremendous political instability. They are all sort of powder kings. And there's an interesting thing
00:25:04
that happened in each of those three countries that I think has the potential to turn the tide. And it speaks to what
00:25:11
Gavin said. Each of those three countries now have banned social media when you're 16 and under. And I think
00:25:18
it's sort of like making sure the kid doesn't get addicted to the drug too early.
00:25:25
And when you look at somebody like Zoran Mdani, I think Gavin is right. I'll go even further. He
00:25:32
AOC, the lady that just won >> Shiovalier. >> Shalier. Let's just break it down.
00:25:39
They're all good-looking. They're all charismatic. They have their pulse on the the gestalt of the moment. They know
00:25:46
how to use social media, and they are essentially curating an army. Now, if you cut the legs off by saying
00:25:55
young people should actually age into social media, I suspect the most important channel of information
00:26:00
consumption being taken away from them actually starts to give them the opportunity
00:26:07
and the rest of us quite honestly because we have to deal with kids who are just really stupid constantly asking
00:26:12
for this stuff. Um, the chance to actually show you how a balanced diet actually allows you to be much healthier
00:26:18
in your later life. And I think as relates to information consumption, I think you are going to see I suspect a
00:26:27
far less radicalized youth aging into the voting roles in those three countries. And I suspect if you start to
00:26:34
see political stability and predictability in Canada, the UK, and Australia, you can put your finger right
00:26:41
on this ban as the reason why. And Florida's already done it in the United States, and I think we need to deeply
00:26:47
consider it across the rest of the United States. Good night. >> Social media ban is now starting at 16
00:26:53
and uh Canada has it. It may come to the US. Go ahead, Travis. >> I'm a full counter to Chimoth on this
00:26:58
one. The I I think we all agree social media is bad for the kids and for the adults. Like too much of that stuff is
00:27:06
very bad. It's brain rot for real. And it's going to be worse than cigarettes. It's all the things. But the real point
00:27:12
of banning under 16 is so that you can force adults to identify themselves and deanonymize themselves so you can set up
00:27:23
a fullscale censorship regime which they're sort of contemplating in the UK. And what censorship is really about is
00:27:30
not about harmful content. It's about content that the people in power don't want you to see that disagrees with
00:27:38
them. if it agreed with them, it's not harmful. It's the stuff that doesn't agree with them. And so what they're
00:27:44
doing is they're criminalizing disagreeing with them. >> And so there's a derivative of this
00:27:51
agebased stuff, >> age gating >> that >> which is really about the adults, not the kids.
00:27:58
>> That is the downside to it. Gavin, you wanted to add? No, I super agree that I
00:28:03
think that a social media ban under 16 would be great, but I think it comes at a really high cost outlined by Travis.
00:28:11
And that's what's really going on is they want to restrict anonymous accounts on X who say things that particularly
00:28:18
the powers that be in the EU do not like. And if there's a way to do that while preserving anonymity and free
00:28:26
speech, I'm all for it. But I think if if it were not for free speech and and X like I think we'd be living in a very
00:28:34
different world today that would be a lot worse. And just to riff on a few things that that all of you have said
00:28:41
you know on communism it is communism and the great tragedy of human experiences we can't learn from the
00:28:47
experiences of others. And it may not matter that communism has failed utterly and ended in death and misery wherever
00:28:57
it has been tried in a variety of different cultures with a variety of different mechanisms and every
00:29:03
generation may need to experiment with it. I think the the saving grace and the importance of preserving free speech is
00:29:11
that the DSA's policies and I would say in particular super progressive democratic policies
00:29:18
are measurably bad. They lead to bad outcomes. If you care about black lives, there's a study
00:29:25
that is uncontested that when you elect a Republican DA, all cause mortality for
00:29:31
young black men in that city drops by 7%. That's all it takes. >> Yeah. >> You care about the environment. Well,
00:29:39
progressives, there's so many regulations that you can't build solar, which is really the only thing that
00:29:44
matters. And like the world is going to run on sunlight. You care about education. Their approach to education
00:29:52
of getting rid of these elite schools that allows low-income people to have a real chance, you know, getting rid of
00:29:59
math in California. It is profoundly disadvantageous to people. It leads to terrible outcomes. And we all know what
00:30:08
happens with crime. It turns out that there's a certain percentage of people who, you know, are well, I think it's
00:30:16
like 60 over 60% of the violence or 75% of the violence comes from people who have like 10 or more convictions.
00:30:24
>> Yeah. I think the specific stat is like.1% commit 70% of the crimes. And if you just dealt with the 0.1%, you'd
00:30:32
effectively no crime. And so I just think the scary thing is and I do worry for the first time like as a student of
00:30:40
history the United States has been like a very stable political economic entity geographically stable for a long time.
00:30:49
And that's to its credit because we could have literally taken over the world after World War II.
00:30:55
But man, if they get total control of some of these cities and drive out everyone who is productive, I don't know
00:31:02
how you come back. And so that's that is a little worrisome to me in terms of the
00:31:08
future of of the United States. >> Well, and by the way, I I I don't think it's just New York. I mean, that race
00:31:16
for mayor in LA where was it? Ramen somehow beat Spencer Pratt thanks to ballot harvesting after voting day or I
00:31:25
should say votes that were found and counted after election day. I mean, that will be a test of the DSA because they
00:31:32
are highly organized and they have learned how to take advantage of all these rules, these ballot harvesting
00:31:37
rules and all these types of things. the DSA, I think they've got something like
00:31:41
half the city council seats now in LA and they're growing. So, especially in these low turnout elections, and look,
00:31:48
this was a Democratic primary in New York, which is a strongly blue state. So, I think maybe 17% turned out some
00:31:56
very low, but this is where the DSA really thrives and excels because they care passionately and they're highly
00:32:03
organized and they know how to take advantage. This is why they want all these like rank choice voting and all
00:32:08
these types of things. they know how to manipulate and take advantage of those kinds of systems. So, I think that
00:32:14
you're going to see this in lots of other jurisdictions wherever they're organized. I think LA will be a really
00:32:19
interesting test. So, I don't think we can just chalk this up to Zoran's popularity. You know, this is a national
00:32:25
movement and we're going to see in a lot of places, but there's no question that
00:32:29
Mandani is now kind of the spiritual leader. I mean a lot of these people don't they think AOC is a sellout you
00:32:35
know or Bernie is a sellout you know they are way more radical than even even those types
00:32:41
>> to add to that sex they did learn something from Trump which is builds a big tent so you're seeing Bernie Mandami
00:32:47
Roana they're all kind of like yeah we're different but there's enough room in this tent
00:32:52
>> well what's happening is that these sort of more established progressive leaders
00:32:57
they want to tap into this energy and even the establishment wing of the Democratic party is bending the knee.
00:33:03
And so what you're going to see is regardless of how many DSA candidates actually get elected, the rest of the
00:33:10
party is now responding to this and they're going to bend. They're going to blow in this direction because they
00:33:15
don't want to get challenged in a primary. I mean, think about this. You had three major congressional races
00:33:20
where the Mamani candidate won. Two of them unseated, you know, really strong incumbents. These were big upsets. So,
00:33:29
you got to think now that every congressional race in a pretty blue district, those members are now going to
00:33:35
have to take into account that they could get primar and they're going to have to tilt their voting and their
00:33:41
views and their rhetoric in this DSA direction cuz they don't want to have happened to them what just happened to
00:33:47
Dan Goldman in the New York 10th district. And just to make one last point on that. So Jal, you mentioned the
00:33:54
the Israel issue and I actually think that is a hugely important and salient issue now in the Democratic party in
00:34:01
Democratic primaries. Obviously you saw that as part of the DSA platform, one of
00:34:05
the legs is free Palestine. This defeat of Dan Goldman, two-time congressman, he
00:34:10
led the impeachment effort against Trump. He had all the right progressive credentials. He checked the box in all
00:34:16
the left-wing policies. He was on the right cable news channels all the time. No one expected him to lose. He lost to
00:34:23
Brand Lander really just over this issue of Israel. Dan Goldman is is very pro-Israel. He basically defended
00:34:31
Israel's actions over the past few years. Whereas Lander, who like Goldman is also Jewish. So this was again, you
00:34:39
know, white Jewish congressman against white Jewish longtime New York politician. So on paper they're very
00:34:46
similar. It was just in this issue of Israel where they disagreed and Lander went to a mosque in order to denounce
00:34:52
what he called the genocide in Gaza. So this was really as close as you can get to a straight up vote on that one issue
00:35:01
in this primary and Lander won pretty handily. Now the reason for this is if you look at polling 80% of Democrats now
00:35:10
say they disapprove of Israel. So you know the approval disapproval rating. Israel used to have high approval
00:35:15
ratings pretty much across the board. I mean, it was sort of a consensus, both Democrats and Republicans. Now, 80% of
00:35:22
Democrats say they disapprove. You really can't underestimate how much of a motivator this is for young Democrats.
00:35:29
They believe >> they were all over campuses. They We see I mean it draws people out and it has
00:35:35
been quite polarizing just to call balls and strikes here inside the Republican party as well. You have Tucker leaving
00:35:40
the party over this. You have Megan Kelly going mental over this and you have, you know, a lot of civil war
00:35:46
inside this administration according to the report. So this is >> Well, let me let me let me um look, I
00:35:51
mean, I'm not taking a side in this. I'm just trying to describe what's what's going on. So
00:35:55
>> absolutely, same as me. >> I think the Republican party is a little bit more mixed on this issue and it
00:36:00
really comes down to age. So if you're part of the older, more establishment Republicans, let's say you're a Fox News
00:36:07
viewer, you still have high approval ratings for Israel. But if you're under 50, which means you're probably not
00:36:13
watching Fox that much anymore. You're probably on to the podcast. I think the the disapproval rating for Israel is now
00:36:19
57%. Interesting. So it really comes down to age. Young people across the board >> have serious problems with what Israel
00:36:25
is doing. And then, you know, as you get into older age groups, that's where you
00:36:30
see a big difference between Republicans. >> I think Chamathy did a wonderful job of
00:36:33
explaining this on Megan Kelly. There's Jewish people, there's the state of Israel, there's Israelis who live in the
00:36:39
state of Israel who are also Jewish. And then there's BB Netanyahu. And I think there's a lot of concern on Netanyahu is
00:36:47
just absolutely out of control is the consensus I think amongst many people across many religions and and political
00:36:55
parties. >> Not to rehash this whole thing, but there are a lot of parallels from what
00:36:58
happened in 107 and what happened in 9/11 in the following ways. When you invade a country and you slaughter their
00:37:08
people, it creates an enormous injury, an enormous emotional, physical, psychological injury.
00:37:18
And what that country typically does is respond by giving the authority to the leader at that time to
00:37:26
set the table right. And you have to remember there was a lot of things that happened post 911 that under any other
00:37:31
circumstance would never have happened. At the top of the list would have been the Patriot Act which in any other world
00:37:36
would never have gotten passed and would have seen the light of day but not for 9/11. So there are moments where leaders
00:37:43
are put in a position and they essentially act on behalf of their country and their people to write a
00:37:48
wrong. I understand that. I think everybody understands that. What's happened now though is that people
00:37:53
cannot logically disambiguate Jews, Israelis, and BB. And I think that's very unfortunate because we're at
00:38:04
a point in time where everything gets conflated and this thing has become this third rail issue. And I find it
00:38:10
absolutely shocking. It's completely reasonable for people to have a point of view on BB and say,
00:38:17
"Hey, you know what? it's enough or it's gotten too far or whatever it is. >> And I think that there's a very
00:38:24
reasonable claim to make that it's time to find different leadership, new leadership inside of Israel and have an
00:38:30
opportunity to reset their standing on the global stage. They deserve that. The Israeli people are incredible. Jews are
00:38:36
incredible. But the idea that you fold it all together and you look at one person and then you apply it to an
00:38:44
entire country and then an entire religion is insane. May I add one thing? May I add one thing is I think there's
00:38:51
well two things really and I'll be quick. >> One, Israel has a giant PR problem. They
00:38:56
need a young sub 35year-old American Israeli who's super fluent in English conversant in social media and
00:39:05
is their kind of spokesperson to America and they need that in France, in Germany, in every country and they do
00:39:11
not have that. There was a young I think his name I forget his name but there was
00:39:14
a young Israeli who was doing a really good job Israeli American representing Israel after October 7th and evidently
00:39:22
BB Netanyahu's wife didn't like him so they canned him and they've never really found a replacement and I do think this
00:39:29
is an urgent issue for Israel because it's like they are taking body blows every second and they're not
00:39:39
even responding. And what I think they do, what they tend to do is they'll sometimes roll out
00:39:46
someone who's a great man or woman in Israel and is in their 60s and a hero of a war, but isn't that fluent in social
00:39:55
media. Maybe the command of English is, you know, technically precise, but there
00:40:00
there's an accident. And those are great people, but they're not great spokes pe
00:40:05
spokespersons. And this is an urgent issue. As far as the antidote, going back to Chamas's comment, I do think a
00:40:11
big reason Trump got elected, every Democrat, many of the Democrats I know who voted for Trump, a big part of it
00:40:19
was that during co they heard what their children were being taught >> and the first time
00:40:26
>> and it was radicalizing for them. >> Yeah. >> And I think it's just really important
00:40:30
that like as part of the American educational system just there not be overtly anti-American things. Listen,
00:40:37
we've made mistakes as a country. We're not perfect, but we're about as good as it gets and we need to tell that story
00:40:47
in every grade consistently and we can have a debate. We've done so many things wrong. Let's learn from
00:40:54
them. But just, you know, slavery was endemic to the world apart from some countries in East Asia. Like, it's not a
00:41:02
uniquely American problem. And we need to tell those stories because I think a lot of kids, the reason they're so
00:41:08
susceptible to this social media propaganda is they've been brought up being ashamed to be American, ashamed of
00:41:15
various things in their identity and being told that America is evil and we need to unwind that because America is
00:41:23
awesome and we're the only vaguely successful multicultural society on planet Earth.
00:41:29
>> Yeah. The melting pot. And if you say that, Gavin, if you say the melting pot
00:41:32
is a beautiful thing, you're going to get cancelled because oh my god, you're getting rid of people's culture. It's
00:41:38
like, no, no, keep your culture and then join this culture and we can all have this great smorgish board.
00:41:43
>> Yeah. You know what your culture is? You know what my culture is? >> I don't know.
00:41:47
>> It's no longer Laura Piana. I know that. I know you're off the train. >> Winning.
00:41:51
>> Winning. The culture of winning. I love it. >> Learning. >> Learning. >> Progress. Adventure. Moth. Do you piss
00:41:58
excellence to quote quote Ricky Bobby? >> Oftentimes I piss excellence >> just burns but oftentimes um just to
00:42:07
warn just to warn the >> dopey Democrats who have been unable to get anything done.
00:42:13
>> The socialists are using your party and uh this guy went viral. They're they're just a host. Like they
00:42:21
literally want to infect the dem Oh yeah, good image. uh they want to infect the Democratic party and then like just
00:42:28
literally get the voting base. And this guy Gustavo Gordillo, I don't know if you guys saw this, he's a DSA co-chair
00:42:33
in New York City. And he just said it outright. I'm going to read you the quote. We're part of the Democratic
00:42:38
Party caucus, but we don't agree with the way the Democratic Party runs its apparatus. So, we're trying to build our
00:42:43
own independence by focusing on volunteer-led movement. We think everyone should be able to be trained
00:42:47
and become someone who can participate in the political process. And we don't think the Democratic Party is run that
00:42:53
way. In terms of the agenda, there's a problem in the Democratic party. They are funded by billionaire donors and at
00:42:58
the same time they're trying to represent the working class. In our opinion, you have to choose between the
00:43:01
billionaire class and the working class. They are just taking over the party. And
00:43:05
again, back to the playbook, this worked really well for Trump. He took the Republican party over and he owned them.
00:43:12
And they tried to get him out. They tried to get him out, but his message and his communication style was just too
00:43:18
on point. He knew exactly what people wanted to hear and he knew exactly how to deliver it. He is a all-time comedic
00:43:26
performer like of non of non- comedians he's the number one comedian in the world and of actual comedians he's in
00:43:33
the top 20 so he communicates perfectly and that's exactly what Mandami's done. He's taken the Trump playbook and he has
00:43:39
applied it here. He is taking over with this communication. You guys know I'm a diehard Nick fan my whole life. cried
00:43:45
and was at game five when they won the finals. And then Mandami gave this speech that somebody wrote for him about
00:43:50
the Knicks and I was just absolutely flabbergasted and upset that it was so great. He's got that Obama Trump
00:43:57
charisma and he is going to destroy your party from the inside out. Socialism is
00:44:02
communism and is the road to suffering and pain. As Gavin said, no good will come out of it.
00:44:08
>> Isn't it amazing? >> Here we go. >> Okay. You know, you know Manny is a communist and what he represents is
00:44:14
basically evil. And yet >> because he just gave a speech about the Knicks, you love him now.
00:44:20
>> No, I love the SPEECH AND I'M LIKE THE SPEECH. >> That's all it takes. We're screwed.
00:44:26
That's all it takes. >> It's true though. If >> I mean this guy, he's a total phony. I
00:44:30
mean, he doesn't stop smiling. He's got this like crocodile smile all the time. He's talking about, you know, eating
00:44:36
you. >> Yeah. >> And it's totally fake. >> Absolutely. I mean, come on. He was he was making Knicks references
00:44:43
and it was the most get out of your seat standing ovation cheer, you know, speech
00:44:48
I've ever heard about the Knicks and I was infuriated. This guy is such a good orator.
00:44:54
>> You got programmed. >> I hate him. >> We're going to have to deprogram you now.
00:44:57
>> No, no, no. It was like getting like uh hypnotized and I just pulled myself out.
00:45:01
I got like pulled in for a second. >> Next topic. >> Next topic for sure. Oh my god, great good strong first
00:45:10
topic. I didn't know you guys were going to go all in, so to speak. Topic two, Chinese open source models appear to be
00:45:15
catching up with the US frontier models. Let's start with a GLM 5.2 released by China Z.AI. This is a Frontier class
00:45:23
open-source free to download anywhere model. 744 billion parameters, 1 million token context window, and it's under the
00:45:31
MIT license. If you don't know what that is, open source uh licenses have very >> super open source. Yeah,
00:45:38
>> it's super open source. Thank you. >> The most open source. >> The most open source of all open source.
00:45:42
If you open it up, you can use it however you like. Uh you can fork it. You can build your own company based on
00:45:47
that. No regional restrictions, no API, fully self-hostable, no uh no Dario. >> It's just yours. It's just making your
00:45:56
chain. It's yours. And you just got to you just got to reference the you just got to reference
00:46:02
the license. That's it. >> Yeah. You shout out the license and you're good. >> Scored 51 points on the artificial
00:46:08
analysis intelligence index. That's the highest score of any open weight model ever. Stacks up nicely next to the
00:46:14
Frontier models. Beat GPT 5.5 on the Frontier SWE coding benchmark. That's a software one. Trails Claude Opus 4.8 by
00:46:22
less than 1 percentage point. API usage cost obviously much cheaper 85% cheaper in fact than GPT 5.5 for comparable
00:46:30
performance Z.AI founder told Elon Musk open weight fable capabilities will be here sooner than Q1 2027.
00:46:38
Gavin, in other words, all this hand ringing, all of these legal restrictions, self-imposed restrictions
00:46:46
are now completely or close to completely moot if they're going to have a model in Q1. Does this 6 months even
00:46:54
matter? Does 6 months in the grand arc of AI matter or not? And what does this mean for Frontier models?
00:47:01
>> I do think how good GLM 5.2 is has challenged some of my beliefs. And there was a great post from a TPU engineer
00:47:10
that for sure distillation has happened. There's been an immense amount of distillation. No question. You know,
00:47:16
>> please explain to the audience what that means. >> Distillation is when you have, you know,
00:47:21
a a like, you know, we all have seen videos of these Chinese iPhone farms. Just picture a farm like that. tens of
00:47:29
thousands of phones, iPads, and computers that are asking the cloud API through masked accounts, very specific
00:47:38
questions, and then these what's called reasoning traces are being harvested because if you're on the API, you know,
00:47:45
you get to see every token. And those reasoning traces are then fed back into the model during the reinforcement
00:47:53
learning process and probably during the pre-training process. And that is a way
00:47:59
that you can get really really close to the frontier at a fraction of the cost. And this is for sure going on and has
00:48:06
gone on for a long time. I do think >> it's a cheat sheet. It's a cheat sheet for other models to catch up.
00:48:12
>> It would be like asking Google every question you could ever imagine, every search imaginable, getting all the
00:48:17
results and then putting your own search engine together to make it very simple.
00:48:22
>> Exactly. But now that this model is so good, it is good enough to do its own RL
00:48:28
and you know the kind of cat may be out of the bag. Now I don't think we really know how good Mythos is. We don't really
00:48:35
know how good the next OpenAI model is. We haven't seen the next SpaceX model. So maybe that gap opens up again. But
00:48:42
either way, I profoundly believe the future is composable models and you're going to every enterprise. You're going
00:48:48
to have a what Andre Karpathy called a council of LLMs. You're going to have, you know, you're going to have Grock.
00:48:54
You're gonna have Anthropic. You're gonna have OpenAI Google. You're gonna have at least two of those. I would
00:48:58
argue Grock should always be one of the two because of its dedication to the truth and it will tell you as a business
00:49:03
owner a politically inconvenient truth that you need to know for your data. But you're also going to have your own
00:49:09
open weights model that you are on your data. And you're gonna put those two together, the frontier models and your
00:49:18
own model and you are going to get you know real paro dominant outcomes and you know half the queries are going to be go
00:49:26
to the open source model maybe 85% and only the hardest ones are then maybe they all go to open source first and
00:49:33
only the hardest ones are then checked by the frontier models. So I think this is the future. It's coming. And a
00:49:41
misconception that a lot of people have is that open- source models are, you know, somehow bad for AI. They're
00:49:47
awesome for the AI infrastructure providers. They just shift economic value from the margins of the frontier
00:49:54
labs to the infrastructure. And that's not bad for AI. >> That's great for them.
00:50:01
>> It's great for them. It's >> great for them. But I do think there's still a role for these frontier models
00:50:06
and it may be true to date frontier tokens are capturing 90% of the economic value and open source tokens are
00:50:13
probably 80% plus of tokens processed and those ratios may be here to stay but I just think composable models are the
00:50:21
future. >> What does that mean composable model? when a composable model where you have
00:50:27
if you're a corporation if you're um do is there is there a new name for your super secret um startup Travis what's
00:50:37
the new name >> the name is called Adams it's not super secret it's super awesome
00:50:42
>> super awesome super awesome um for your super awesome startup Adams which man
00:50:48
lot loads of people have been calling me to say how like they've been in the you
00:50:52
know they they've spoken to you and >> invest investor the uh your your dog is hunting with investors, Travis.
00:50:59
>> All right. Fair enough. Fair enough. >> Okay. >> You're going to have a router and every
00:51:04
query that somebody comes in, every task that needs to be done at your company, that router is going to send it to, you
00:51:10
know, your RL SFTD version of Dematron. >> Yes. Then at some point in the workflow,
00:51:18
a frontier model may or may not come in to kind of check it, add to it. And that's what I mean by a composable model
00:51:27
when you have kind of um, you know, kind of a symphony of models working together
00:51:31
with kind of the frontier models being, you know, maybe the conductors. >> But that's what I mean when I say
00:51:36
composable. >> Yeah, understood. Thank you. >> Saxs, formerly AISAR and and now running pcast. What are your
00:51:45
thoughts on China's ascension in open source? And we're still looking for our open source champion obviously here in
00:51:52
the US, but feels like >> Can I just say one thing? Nvidia is the American open source champion.
00:51:57
>> Thank you. Yeah, >> they can release GLM 5.2 or better whenever they want. >> Okay. But why haven't they done that?
00:52:06
They don't want to screw their customers too much of a conflict to be like pushing it out there too often.
00:52:11
>> All of the above. Who knows? We'll see. Okay. Yeah. And you got to be delicate
00:52:16
there. I see. Okay. So, I'll say it, not you. Jensen does it. Yeah. I mean, it's
00:52:20
it's a it's a classic channel conflict, right? You're you don't want to compete with your customers. And he is competing
00:52:26
with Elon on self-driving now. And Elon's making his own chips. So, >> I think these frontier model companies
00:52:32
should think very, very carefully about A6 and the incentive that creates for Nvidia. On the model perspective,
00:52:40
>> if OpenAI, which launched their Jalapeno chip this week and announced it being
00:52:45
built by I believe Broadcom and they are saying, "Hey, f you to Jensen and Nvidia
00:52:51
and they already were full contact with him." Don't be surprised if Nvidia says,
00:52:56
"You know what? We kind of like the area you're operating in now that you're going to make chips and maybe OpenA
00:53:02
sells those chips to other people." Don't be surprised if Nvidia starts an OpenAI competitor. You heard it here
00:53:08
first on allin. Sax, did you want to jump in there? >> Sure. I mean, look, I think that China's
00:53:14
been good at open source for a while here. There's nothing new about that, but there are a few things that are
00:53:20
significant about GLM 5.2. So, the first one is it is now like you said the best
00:53:25
openweight model for coding, software engineering, and long context agent work. And you gave a couple of the sweet
00:53:34
bench scores. I mean it was just a tick below OPUS 4.8 and it was right up there
00:53:39
with GBT 5.5. So if you compare this to again the state-of-the-art the frontier models for anthropic and open AI it is
00:53:50
right there with the previous model. But you have to remember now that the current model fable for anthropic and
00:53:59
5.6 six for open AI is now in a little bit of a purgatory because of all the reasons we covered last week. Now look,
00:54:09
I like I said last week, I ultimately blamed Daario and the way he communicated and the way that he primed
00:54:15
officials to be on a hair trigger with respect to these models and when the government got a credible report about a
00:54:23
jailbreak from, you know, one of Anthropic's own most trusted partners, you're going to say roll that back. But
00:54:29
that is the situation we're in right now is that Fable has been rolled back and GBT 5.6 is trying to navigate these new
00:54:37
approval hoops. So we now have a Chinese openweight model that is as good as the
00:54:45
currently available models from OpenAI and Anthropic. And look, this is a point I've been making really since I joined
00:54:53
the administration is that we are in a very competitive situation with China. I've been saying this from the
00:55:00
beginning. Our whole AI strategy from the get- go was about winning this AI race, defining it as a race, as being
00:55:07
globally competitive. And we cannot afford to do things unnecessarily that slow our companies down. I hope that
00:55:16
>> David can ask, >> do you think Daario got exactly what he wanted? It seems to me there's some
00:55:21
chance, this has been a very calculated strategy to provoke the US government into doing what they just did and this
00:55:27
is what he wants. He has a regulatory moat now. He can keep his future models behind this, you know, give it out to
00:55:34
Glass Wing, use it to distill it for him for themselves. Do you think this is what they wanted? I think that on a
00:55:41
certain level it is what they wanted because they've been advocating to have a federal regulator, basically a new
00:55:47
agency. In fact, Daario posted a blog just a few weeks ago saying he wants an FAA for AI. They wanted government
00:55:52
approval, a government approval process for AI models. And so, in a sense, they've gotten exactly what they wanted.
00:55:58
Now, that being said, I don't think they're happy about the fact that Fable has been rolled back. So, in a sense,
00:56:07
you could say that Daario got hoisted on his own petard here, or it could be a FAFO situation.
00:56:15
But look, my view on it is we should not reward Daario by giving him exactly what
00:56:20
he's always craved, which is some sort of labyrinthine government approval process that does reward regulatory
00:56:26
capture. So I hope that very soon now I do think that as long as Enthropic has resolved the jailbreak issue then I do
00:56:38
think they should be allowed to come back to market and similarly for open AI I don't think we should be delaying them
00:56:43
unnecessarily. We do not have months to give away in this race. And let me just say one other thing which again is
00:56:50
something I've been saying for months which is with respect to risks like cyber it is undoubtly a risk. But what
00:56:57
is the response to that? The only thing you can do is go out and find all the vulnerabilities first yourself. The
00:57:04
white hats have the white hats find all the vulnerabilities and do a big upgrade
00:57:08
cycle. Roll out the patches before they can be exploited. The reality is that if
00:57:13
you just clamp down in a way that doesn't even allow these models to be used, the Chinese are going to have
00:57:21
these capabilities imminently anyway. You know, they're already at Opus 4.8 level. And the founder of Z.AI, he said
00:57:29
that before Q1, they'll have fable level capability. I believe them because look,
00:57:34
the the Chinese have been, I'd say, 9 months behind our models, plus or minus 3 months depending on capability. But
00:57:41
it's when they know there's been a breakthrough around something like cyber, they can deploy more resources
00:57:47
against that particular problem and catch up faster. >> So, like I said, we're on a shock clock
00:57:52
here. I've been saying for months that we're on a shock clock. We have to do smart things. We can't just slow
00:57:57
everything down because that will not slow down the Chinese. They're not under our jurisdiction. We have to basically
00:58:04
get these tools in the hands of our cyber security industry. They're the force multiplier. They're the enabler.
00:58:10
We have to basically go out and do this big upgrade cycle quickly because we only have a few months left.
00:58:16
>> They're six months behind on the model and they're 24 months behind on the silicon yet they're only a few months
00:58:21
behind in total. So what game are we playing? This is insane. We are going to lose if we keep doing this stuff to
00:58:28
ourselves. >> So let me make a point about that. So on the silicon, there's been a huge push in
00:58:33
China by the government to push their AI labs to develop and train on Huawei chips. And look, you can take these
00:58:42
claims with a grain of salt. Maybe they're not true, but it was claimed that Deepseek V4 was trained on Huawei
00:58:48
chips. And now Z.AI, they are saying that the GLM5 family was trained entirely on clusters of Huawei Ascend
00:58:57
910b chips. So now look, maybe they're lying, maybe they smuggled in some Nvidia chips, but the claim is that this
00:59:05
was all done on indigenous chips. And what I believe is that China is engaged in a strong indigenization push right
00:59:14
now. They want to prop up Huawei as the national champion. They want all their companies using Huawei chips. They still
00:59:21
need to scale some of the manufacturing, but they're going to do that pretty quickly. And then what they're going to
00:59:26
do is they're going to take these Huawei chips. I'm going to take these Huawei optimized models. Remember that
00:59:33
GLM 5.2 the inference is optimized for the Huawei chips. Okay, we know that. And they are basically going to package
00:59:42
these things up. They call it AI in a box. They're going to sell it at a fraction of the cost globally,
00:59:47
>> which is what they do with every technology, right? Better, cheaper, faster >> or almost as good. Yeah. And that's
00:59:53
another thing, as I've been saying since the beginning of this administration, we
00:59:55
have to be proexport because China is going to be there within a one or two years. As I said, we're going to be
01:00:01
kicking ourselves because we could have had the whole global market to ourselves. We invented reasons not to
01:00:07
sell abroad to our friends and partners. And now China is going to be there imminently.
01:00:12
>> Yeah. And with a lower price seg when you play with uh this new GL, you get some really interesting responses. it uh
01:00:20
I asked it about the country of Taiwan, was not pleased and uh didn't give me an
01:00:26
answer. I asked it about Tiana Square. No answer as well. I'm using the hosted version at Z.AI. Uh but when I asked it
01:00:32
places to visit in Paris, it did an exceptional job. Except when I said make these ideas into an infographic and make
01:00:38
me like a 3-day agenda, it was like, hey, we don't have enough time to do that. And it said use the other model
01:00:44
because this model's too busy, but you can go play with it at z.ai. dish on a point about the censorship. So
01:00:51
there's no question that these Chinese models have you could say censorship and you know there's political bias in there
01:00:59
out of the box. But American companies have taken Chinese models and then essentially worked around and basically
01:01:07
fixed the censorship inside their own forked version. So for example, Perplexity did this very early on with
01:01:14
Chinese models. they showed that you could sort of put back the content on TNM Square and things like that. So I
01:01:20
think Jal, you're absolutely right about the censorship, but it's not a fatal problem. It's something that American
01:01:26
companies can fix when they >> Yeah. >> take an open source model and fork it and customize it.
01:01:31
>> Yeah, in the hosted version, you're not going to get a great answer of what agenda you should use for tourism in the
01:01:39
great country of Taiwan or your visit to Tanaman Square. All right, let's keep moving here on the dock at Micron Smash
01:01:47
their earnings. If you don't know Micron, they are one of only three companies on Earth that
01:01:52
make high bandwidth memory. These are specialized chips. They sit on top of the Nvidia GPU and their entire 2026
01:02:00
supply is sold out and has been for some time. SK Heinix and Samsung also make HBM. Micron smashed earnings. revenue up
01:02:08
4x 4x year-over-year 9 billion to 42 billion beat expectations by 16% big jump in guidance for Q4 50 billion
01:02:18
versus 43 billion their stock is up 10x shout out to Gavin in our 2025 prediction show he gave uh a call on HBM
01:02:28
makers like Micron as the best performing asset since that time up 14x >> I'm not crying in my soup
01:02:35
>> you're not crying in your soup I got a ton of information here. I think this I'll just end on the Apple price
01:02:39
increases. Everybody knows Apple has been really uh been a beneficiary of the run local models movement that I'm part
01:02:49
of and and people are buying 128 gig, 256 gig MacBook Pros, Mac Studios. But the gig is up apparently because now
01:03:00
Apple, which had not passed on those costs to customers, is having to pass those increases on. So everything from,
01:03:08
you know, the new MacBook Neo, which is their $699 laptop, you know, kind of competing with
01:03:14
Chromebooks, is now $7.99, up 15 14% and Mac Studio up 25%. The costs are just going to be very significant. Inflation
01:03:24
has come to the desktop. Your thoughts, Gavin, on Micron and the impact on the industry, and is this a temporary
01:03:31
bottleneck or does this mean everybody has to get into this business quickly? No. Well, one DRAM is the most important
01:03:37
bottleneck. There's a whole segment of people on X who are very focused on bottleneck. I bottlenecks. I call them
01:03:43
the bottleneck bros. You know, they'll they'll do some work with Claude, find some esoteric Japanese company. The
01:03:49
bottleneck that matters is DRAM. And DRAM and HBMD RAM, this is the most important bottleneck simply because
01:03:56
memory capacity and bandwidth are foundational to the performance of every AI model. So, this is the most important
01:04:02
bottleneck. Elon is focusing the terrafab on memory because he sees it as the most important bottleneck. You know,
01:04:08
not lasers, not capacitors, not power power supply semiconductors, not NAND flash, not HDDs, DRAM. Uh, and I think
01:04:18
this bottleneck is going to be with us for a while and it is kind of astonishing. So I think so a few
01:04:24
thoughts like what was important about the quarter they announced that they have these SDAs these supply chain
01:04:28
agreements that have a floor and a ceiling for prices with increasingly large group of large customers and this
01:04:40
covers essentially 50% of their revenue I think with just four customers and the
01:04:44
floor pricing in these new contracts is ahead of prior cycle peaks from a gross margin perspective and so this is
01:04:55
really I think pretty maybe end up being very transformational for the industry.
01:05:02
most other parts of the semiconductor supply chain have rerated you know lamb research
01:05:10
you know the the wafer fab equipment suppliers you know they all trade at huge premiums to DRAM relative to prior
01:05:17
cycles and their business models have improved but you know so has the industry structure and business models
01:05:22
of DRAM because HBM DRAM is increasingly a customized chip but as far as other people being able to do this look CXMT
01:05:31
is going public in China. They are going to they may be the cure for Apple's ills. They will flood the market with to
01:05:39
some degree cheap consumer grade DRAM, but for the DRAM you need in these AI servers, there are three companies that
01:05:47
can make it. It's really hard to do. This is as close to magic as science can get. And you know, I think Terraab
01:05:55
Terapab, you know, is going to be an important part of this solution. But um you know these these stocks still trade
01:06:02
are cross-sectionally cheap relative to the rest of AI. Something I've been thinking about memory is DRAM is
01:06:08
probably going to be 30 to 40% of all hyperscaler capex next year. Every do the hundreds of billions of dollars that
01:06:16
are spent going straight to DRAM. It's wild. But this may actually be very valuable for
01:06:24
society because it is probably, you know, going to, you know, inflate the costs of building a gigawatt data center
01:06:31
to the point where like, you know, even for the hyperscalers, um, economics matter. We're caught in this prisoner's
01:06:38
dilemma. And this may give us as a society time to adapt to adapt, you know, what our friend Brad Gersonner
01:06:45
calls the social contract. So the high iPhone prices, you know, one, CXMT is coming for consumer grade DRAM, but two,
01:06:54
this may be good for AI. It may be good for us as a society >> and making it is just really pure
01:07:03
silicon, right? Like making memory is just incredibly refined silicon >> and that might be the pre- bottleneck.
01:07:11
>> Yeah. Yeah. making the HBMD DRAM, making what Nvidia calls SOCAM, making LPDDR.
01:07:17
These are the types of DRAM that are really hard to make, not consumer grade DRAM, and they are increasingly what you
01:07:24
need in these AI data centers. >> Yeah. Semiconductor grade. Yeah. >> Yeah. So, my understanding of HBM stands
01:07:31
for high bandwidth memory. Again, this is part of the, you know, the GPUs that go in the data center to run AI
01:07:39
is that you take the the DRAM wafer or die and you actually stack them. And so I think HBM 3 is like eight. It's
01:07:49
stacked eight dyes high, but now they're increasing to 12 and even 16. And to basically stack them and then package
01:07:56
them all together is that's an advanced technology in and of itself. So you're seeing now, like Gavin's saying, there's
01:08:03
only three companies that can do it. But also, this is creating significant price
01:08:06
pressure for all the consumer electronics businesses. Apple had huge news today where they announced massive
01:08:12
price increases. And again, it's because DRAM now is less available because it's
01:08:17
just being hoovered up by all the data centers. And if you're a data center and you need to buy GPUs, again, those
01:08:23
chips, they're using immense amounts of DRAM because again, one HBM chip is using multiple like stacks of DRAM. So,
01:08:34
it's just getting slurped up and then it takes a couple years to ramp up new capacity. So, these companies are going
01:08:40
to do that, but that could take a while. We saw that in New York. remember that was that Micron plant that was had just
01:08:46
broken ground and then got shut down the same day because of some crazy environmental issue. So, it's not easy
01:08:51
to ramp this stuff in the US. Although Micron is the one provider that's in the US, SKH is in South Korea. Samsung's in
01:08:59
South Korea, too. But anyway, we're going to see again more of this AIL they're calling it. You know, it's just
01:09:06
another reason I hate AI is it is in this narrow area of consumer electronics where there's competition for DRAM. It
01:09:13
is leading to price inflation. Now, >> Microsoft raised the price of the Xbox. You know, it's coming for the Switch.
01:09:19
It's coming for the PlayStation. You know, there's demand destruction because of prices in consumer whereas AI demand
01:09:25
is relatively price insensitive. David, I would modify one statement. It's hard to build a micron. It's hard to build a
01:09:32
new fab in a deep blue state. You can build fast. >> Why were they trying in New York? It's
01:09:37
kind of crazy. >> You know, New York gave them all these incentives and but it none of those
01:09:42
incentives matter. It's a little bit like solar power. You can be as pro- environmental as you want. But if you
01:09:47
can't build and install solar because of regulations, it doesn't matter. So, you
01:09:52
know, maybe that Micron plant ends up getting, you know, built in my home state of Texas.
01:09:57
>> The incentive game has kind of flip-flopped. It's like it used to be the states were courting the factories
01:10:03
and the fabs. Now it's the fabs are like, "Which state can actually build this? We'll pay you whatever you want.
01:10:08
Just tell us where to send the envelope. We'll we'll we'll drop a couple of envelopes off. It's not a problem."
01:10:15
>> Gavin, can you say how long it's going to take to stand up the fab at Terraab?
01:10:23
>> Well, I mean, if it was a normal fab, it would be a two, three, three and a half year
01:10:29
process. But, you know, we we we've seen what Elon has done to other construction
01:10:34
processes and, you know, he's starting with some some advantages with the Intel partnership. So, I don't think anyone
01:10:42
knows, but based on past history, he's probably going to stand up terapab faster than other fabs have been stood
01:10:49
up. But it what this is really hard. It's really hard. It's the intersection of magic and science. You can't believe
01:10:55
how complicated this is. So, it's going to be hard. But, you know, he has a he has a track record of doing, you know,
01:11:02
what Jensen called impossible superhuman. And so, we'll see. We'll see how long it takes.
01:11:08
>> You know, one other point here that I guess is it might be relevant to SpaceX
01:11:12
AI, although it's don't have to limit it to this is um I think there's an assumption that over time it would get
01:11:19
cheaper and easier to stand up new data centers, right? But what you're saying is actually it might be getting harder.
01:11:26
It might be getting more expensive, right? Because there's competition for these components. The memory is getting
01:11:31
more expensive. I'm not sure that the GPUs are getting any cheaper. I guess some of these
01:11:35
>> transformers, the switch gear, >> the energy might be getting cheaper and then the entitlements are getting harder
01:11:40
and the political situation is getting harder. There's very few places you can even stand up new data centers. So, is
01:11:47
it the case that actually it's going to get more and more expensive to >> 100%. So to stand up a 1 gigawatt data
01:11:55
center, it's $35 billion in semiconductors, Nvidia semiconductors, and it's $25 billion of power and
01:12:01
cooling equipment. And that is clearly inflationary because a lot of that 25 billion is the human labor required to
01:12:07
install it. So the calculation that needs to be done for orbital compute is it's 35 billion of silicon in each space
01:12:15
and you know in in literally outer space and orbit and on ground. But if you can
01:12:20
get the cost of launch significantly below that $25 billion, then the math starts to really mass. And when Starship
01:12:28
is reusable, it's going to cost $5 billion to put a gigawatt of compute into space. And something that drives me
01:12:33
crazy is people picture these Pentagoniz data centers. No, it's racks in space linked with lasers. It's it's kind of a
01:12:41
virtual data center in space. >> Wait, five billion. Is that five billion of launch cost or what?
01:12:46
>> Five billion of launch cost. Now you're at 40 billion to put the gig into space.
01:12:50
You're at 60 billion terrestrially. And the 25 billion that is power and compute
01:12:54
is clearly inflationary. And so it may be that in three or four years it's 70 billion verse 40 billion. And that five
01:13:04
as starship becomes rapidly reusable is likely deflationary. So this is the economics that underpin orbital compute
01:13:12
from first principles. And then on an ongoing basis, you are you're maybe paying a billion dollars a year for the
01:13:19
power to run those chips and cool them. >> If I had to make a guess, I think what's
01:13:24
going to happen is that since 2021 about 40% of all data centers get contested, right?
01:13:32
I think that number is going to go up. So saxs I suspect that whatever forecasted energy consumption that we
01:13:39
are looking at in AI is grossly imbalanced. There is very very meager supply and there's effectively infinite
01:13:48
demand. So that probably pulls forward the economic equation to want to go to space. But then again, that's going to
01:13:57
prefer SpaceX and their compute stack and their compute decisions over the hyperscalers and over anybody else. And
01:14:06
so you're going to have a cost of an output token, I think, terrestrially, particularly from the hyperscalers, be a
01:14:11
little economically lopsided versus SpaceX. Once they get it to scale, now that's the key statement.
01:14:17
Whatever is left on the ground, though, will be incredibly incredibly valuable. It'll be a diamond. These are diamonds.
01:14:23
And the thing is, you have to find reasonable size, right? You can't have a 10 kilowatt diamond. That's like a
01:14:30
little pebble of nobody cares. But if you're in the reasonable hundreds of megawws to gigawatts, man, those are
01:14:37
like hope diamonds. >> Those are just lock them down, >> which I own. Travis, there was a uh interesting
01:14:46
trademark filed this week. So, more in the investigator uh investor uh investigator
01:14:54
uh of the Tesla plus SpaceX marriage that everyone seems to believe is going to happen uh shortly.
01:15:01
>> You can you can you can say that that's you can give me credit for it. >> Of course. Yes. Uh as Chimath has
01:15:07
architected in his uh in his high perch. Uh but this trademark came out. >> How cool would that be? How cool would
01:15:14
it be for those two to come together? >> It's gonna happening. And uh it'll be incredible. And uh if you're lucky
01:15:20
enough to be an owner of both >> top bottom, everybody can decide to be an owner of
01:15:28
>> Jason. Jason, I'll be I'll I'll be making love to myself when this happens. There'll be no
01:15:32
>> So no different than any other Thursday. >> Exactly. >> On the top and the bottom.
01:15:37
>> The top and the bottom just like any other Saturday night. >> Here's the trademark for Megapod that
01:15:43
came out. This is a uh filing date of 618 20226. So a very recent June 18th. Modular data center hardware for
01:15:52
artificial intelligence computing comprised of network of computer servers computer hardware for artificial
01:15:57
intelligence processing computer network hardware electric power distribution units and cooling systems sold as a unit
01:16:02
self-contained modular computing hardware systems for artificial intelligent workloads yada yada yada.
01:16:08
Essentially >> just to explain what this is. Well, yeah. Let me let me just, >> by the way, that description makes I
01:16:14
don't know what it I still don't know what it is. >> Essentially, what people are saying is
01:16:16
this is going to be a giant battery pack with GPUs at the supercharger stations which have already been approved. That's
01:16:24
the back channel. Chamatha, go ahead. >> You have a couple of issues right now to
01:16:28
turn on compute terrestrially. So, assume you have land, that's relatively straightforward. Assuming you can get it
01:16:35
zoned, less straightforward. Assuming you can get power, very difficult. then you have a very critical design
01:16:40
decision. So all of these folks publish these things called the basis of design and your your bods essentially tell you
01:16:47
here's the anthropic spec, here's the open AI spec, here's the coreweave spec, here's the AWS spec, here's GCP and you
01:16:53
get these 50 and up to upwards of 500page documents of all these technical details. The issue that we have is I
01:17:01
don't know if you guys have used OpenAI or Anthropic recently where you get the whole thing of like come back later,
01:17:06
right? That come back later is completely unacceptable. It just means that they have no compute.
01:17:11
So in trying to find >> surge pricing is coming. >> Yeah, let's go. >> So in trying to find this solution, what
01:17:18
is happening is these basis of design the restrictions, the specificity is being relaxed and one of the key
01:17:25
constraints is we've moved to an architectural model from Google and Nvidia that has said look we have to
01:17:32
liquid cool these racks. These are very complicated big girth supercomputer racks essentially. And now we're also
01:17:40
going back and saying, you know what, maybe some of these older stuff that's a little bit less proficient and a little
01:17:45
bit less useful, but we can air cool them is useful. And everybody's like, yeah, you know what? We should try to
01:17:52
use everything that's available in that second class. What some very smart people are doing are like, "Wow, well,
01:17:58
here's a shipping container that you can just drop on a concrete pad somewhere, plug in the power, and let it rip." And
01:18:04
so, Jason, what you're seeing is that level of investment that's happening. So there are companies, you know, like Dell
01:18:10
makes these racks, companies like Vertive makes these modules and I think if Tesla can make them available,
01:18:18
folks like us for my data center project, we would be enormous buyers of these things because we would just
01:18:24
literally prefab them in a warehouse, right? >> Get the chips, prefab them, truck them
01:18:30
to the place, crane them in, turn it on, off to the races you go. you have like a
01:18:35
90day build cycle which is unheard of. So that that's where these mega pods are coming from. I hope it's not entirely
01:18:44
consumed by Tesla and SpaceX internally. >> Well, that's the rumor is that he's going to put them at the supercharging
01:18:50
network where he has a lot of land and he's got power there already and those are lightly used in some cases. You
01:18:55
know, >> there are other requirements. The problem with some of these things is that you need to have certain levels of
01:18:59
access. There are certain infosc requirements. There are certain liquid cooling requirements that make many of
01:19:05
the workload applications unfeasible at a place like a supercharger center where
01:19:09
random bumble are like trapesing around >> security guard put it off to us. So guys, I've got I've got 500 properties
01:19:17
with lots of energy, lots of mechanical cooling systems and lots of uh LNG access or sorry not
01:19:27
LG natural gas, sorry, already piped in. So on all the stuff I'm doing on robotics, AI, physical AI,
01:19:37
we're literally looking at putting some of our compute >> Wow. >> into our kitchens.
01:19:43
compute kitchens. >> You're going to be able to buy small modular let's put data center in a quote.
01:19:51
>> Yeah, but but it doesn't it doesn't matter. Like I I just go buy the GPUs and I'll just I just set it up in a
01:19:57
kitchen, >> right? It's not that No, hold on. It's not a Bitcoin miner. >> No, no, hold on, hold on, hold on. It's
01:20:04
not thating easy. Okay. Like, if you guys want to build a cloud and contribute it to a pool,
01:20:08
>> you're going to have to sign up for liquidated damages. And you're not gonna sign up to
01:20:12
>> SLA. I'm not a hyperscaler. I'm just using it for myself. >> Yeah. Rack and stack.
01:20:16
>> Oh, okay. Sure. You can use it for yourself. I'm saying the more interesting opportunity is when
01:20:22
>> whatever Travis says. Let's talk about the interesting stuff. >> No, no. I'm just saying like
01:20:27
>> what I thought you were going with this, Travis, is like you could build a synthetic pool and contribute it to a
01:20:32
neocaler. And all I'm saying is to do that is a leap of things that you'd have to do that you probably don't want to do
01:20:37
because it >> veers you away from what you're >> doing. I think the security thing is the
01:20:41
main issue around non data center type real estate to be honest. It's the big one. Look, you you got to do liquid
01:20:49
cooling, you got to do mechanical systems for air cooling. Uh you have energy. Like there's a lot of things you
01:20:54
got to do. But assuming you even had all that, the architecture of a data center
01:20:58
is set up like they have man traps, right? where you go into a room, the door closes behind you that you got into
01:21:06
with your fingerprint. Then you go into the next part of it again with your fingerprint only when that door closes.
01:21:12
Like, >> well, just out of genuine curiosity, how much does one of those man traps cost?
01:21:16
Like, why can't you put some of those in your >> Yeah, maybe you could. >> Are you interested in putting in a man?
01:21:20
>> I'm just trying to help trap some men. I'll try to interpret where Chimat's going a little bit is like a mega pod
01:21:27
that's sitting at a char like a public charging situation because of the physical access of it like there's
01:21:34
probably some stuff you've got to do if you're going to resell it but if you're going to use it for yourself you can use
01:21:39
it for yourself the real value I think in the Travis example which I find super exciting is if you can contribute it to
01:21:46
a distributed training pool or a distributed inference pool that has a lower SLA which is more of how this
01:21:53
community-based oriented work. Jason, I think you've talked about Bit Tensor. I think Venice is a project. I think
01:22:00
Plurales is another project. Travis, that's where I think it's super exciting where
01:22:03
>> but but the one thing I'll just throw out there, guys, the distributed training stuff.
01:22:09
If these things are far away from each other, meaning not right next to each other, they're so less efficient. Like
01:22:15
the efficiency drops dramatically. You want these things to be right next to each other physically. You get like
01:22:23
multiple orders of mag at least an order of magnitude type efficiency maybe plus+
01:22:27
plus like it's a big deal. Yeah. >> So like you could have two we have two facilities passing the jobs from one GPU
01:22:34
to the other is super important and if it's on a peer-to-peer network >> it's going to have lag. It's
01:22:39
>> honestly even if it's on your own fiber connected like 2 kilometers away you're
01:22:45
screwed. It's not a thing. >> This is 100% true. But everything that cuts against training works for
01:22:53
inference >> where respect >> well latency is super important and I do think you know distributed inference
01:23:02
distributed inference clouds are coming. >> Yes, I get that. That's >> and to riff and to riff on like what
01:23:07
Chamas said in all of this like one I mean there is actually a startup that is trying to put four GPU units with kind
01:23:14
of a battery on people's houses and give them a discount on their power and then
01:23:20
you can do inference for that neighborhood you know from those four GPUs and it's like lock sealed so nobody
01:23:27
can get in but there's there's another dynamic that I think we should talk about with all of this and can play into
01:23:33
the megapods And you know other people are kind of working on data centers. You know Crusoe is working on you know
01:23:40
modularly assembling data centers you know kind of like a data center and think of it as like an 18-wheeler um
01:23:45
what do you call those things the 18-wheeler you know shipping container whatever it
01:23:50
is. Um but that is the disagregation of inference into prefill and decode. When you when a model is answering your
01:23:57
question it's doing two things. The prefill part is understanding the question and its answer thus far. And
01:24:03
think of think of that as the more you can remember the bigger your you know your memory capacity literally the more
01:24:09
words you can remember the better. Decode is the process of generating the next token and that is a memory
01:24:17
bandwidthbound problem and think of it as the faster you can speak the better and these two types of inference are
01:24:26
increasingly being disagregated and Chimath was an investor in Grock which Nvidia
01:24:32
bought and they're going to use this cerebrus is the other solution today that is available and you can put Grock
01:24:40
or Cerebrris decode, you know, optimized chips. They both say you can do more than decode on them, and
01:24:46
that's true. In front of old Nvidia GPUs like H100s. So, you can you can lift H100's, A100s out of some old data
01:24:55
center, put them in one of these mega mega pods, you know, a rack in a shipping container, put a Grock or
01:25:01
Sirius in front of it, and you can get a very competitive solution. And so I do think the disagregation of inference,
01:25:08
we're going to be using GPUs for seven years, 10 years, 12 years, and that's great because it lowers the cost to
01:25:13
finance them, which makes this AI revolution more financable. Jamath, you want to riff on on Grock? Should we
01:25:20
should we should we >> I really agree with everything. You're incredibly well steeped in the space.
01:25:24
It's so exciting. I don't have any investments or anything in the distributed compute space, but at the
01:25:29
intersection of competing against China, having a vibrant American open source community, having a bunch of distributed
01:25:36
models for purposes of free speech and otherwise that Travis mentioned earlier, I do think this idea of distributed
01:25:42
inference has a real place in the American ecosystem. I don't exactly know where and how and how homeowners would
01:25:49
get paid, but whoever figures that out as a pan-American idea, I think is a is really onto some
01:25:56
>> and yes, Travis, obviously slower, but people are contributing compute to this
01:26:01
that's like kind of surplus computer or comput that's available to >> unused computing unused comput. Uh, and
01:26:08
then there's Targon, which is just straight up people are putting and you can rent H200's by the hour for three
01:26:15
bucks, four bucks, and it's permissionless. Anybody can contribute provided, and this is kind of the magic
01:26:21
of Bit Tensor is there are validators that make sure you're putting in what you're say you're putting into the
01:26:25
network. And so if somebody just had the hardware and Tesla said from now on every Power Wall, and I I I don't put it
01:26:32
past him to do this. You buy a Power Wall, we give you a discount on it. Every Power Wall has our GPUs in it.
01:26:39
Part of the offering is you can't buy a Power Wall without putting GPUs in it. And when you're not using it, we will
01:26:45
pay you for your battery. And you put a Starlink on your roof. And now you've got a distributed system where you get a
01:26:52
couple of of these um power walls with their own silicon in it and a Starlink. And now Elon's created an infinite
01:27:01
number of home battery backup systems. people get their battery system for free.
01:27:06
>> He gets the exclusive for 20 years. >> You guys see that there's a rumor that
01:27:11
Elon's going to buy T-Mobile. >> By the way, we should just because Cerebrus has blown up and gone through
01:27:18
deal price. We should talk about Cerebras a little. >> So, uh, Cerebras had an incredible IPO.
01:27:23
Give us an update on where these IPOs are happening. Obviously, SpaceX had an incredible IPO, but has retreated from
01:27:28
this, you know, otherworldly $200 share price. So, so to the extent you can talk
01:27:32
about these two as well as the two IPOs to come, $4 trillion in backlog. We've obviously got OpenAI and Claude, I'm
01:27:40
sorry, Anthropic, which makes Claude. Both of those will be worth a billy plus. You put all this together, 4
01:27:45
trillion of new offerings plus Cerebras in there in the mix. How does the market
01:27:50
manage this much new inventory being put on the market? Obviously the flow to SpaceX is notably small but over time
01:27:57
people like yourselves and and other insiders founders fund etc will be unlocked uh and have the ability to
01:28:04
distribute to their LPs. So this is going to be a moving target I think on terms of share price and can the market
01:28:10
absorb this? Where does the money come from retail or does it come from people selling their Bitcoin and moving it over
01:28:16
to something more exciting? What's the dynamic here in the market? I know we're in uncharted territory, Gavin. Well, for
01:28:22
sure there's no precedent for any of this, but a few things I would say like just in no particular order. I think
01:28:27
Anthropic is worth $3 trillion today and it's very important. >> I'm sorry, did you say Anthropic is
01:28:32
worth $3 trillion? >> Yeah, I think that is roughly where it would probably trade as a public company
01:28:37
and >> Wow. >> Yeah. >> I mean, look, they're going to do they're going to end this year
01:28:44
>> Oh, man. >> They're going to end this year well over 100 billion. >> Holy [ __ ]
01:28:49
What's the 28 number? >> What's the 28 number? Is it 200? Is it 300 billion? It's probably not going to
01:28:56
trade at 10 times that number and it will be very profitable at that scale because it'll be inference
01:29:04
dominated and people reporting they have 85% gross margins on inference. But in terms of the market absorbing
01:29:10
this like the market's already absorbed it, you know, it's it's just shifting from private to public. And so in the
01:29:17
scale of global capital markets, these seem like really big numbers. You're just moving from the private markets to
01:29:23
the public markets which are even bigger. As far as SpaceX specifically, I think one of the more important things
01:29:29
um is everybody who's a SpaceX investor or employee has had a chance to sell every 6 months for the last 10 years. So
01:29:38
they not may not be the wall of liquidity that some people are thinking about. I read this New York hedge fund
01:29:43
short short report that you could just short SpaceX on the lockup because so many people are going to sell. Really?
01:29:48
Well, everybody who's on the cap table, they had an opportunity to sell and almost half the employees at SpaceX
01:29:57
bought on the IPO. Now, I do think Cerebrus, >> but Gavin, one thing though, I got to
01:30:03
say, so I I I've had SpaceX since 2018. What? What? But their little their little liquidity
01:30:11
thing every year, like last year it was like 350 350 billion last year. So you got an 8x in one year. You could have a
01:30:19
lot of people selling, right? They were doing little 20% up, 30% up clips for many years. And then an 8xer could
01:30:29
create that liquidity. >> Maybe, but we'll see. That's possible. But the employees are buying at the new
01:30:35
price. and they're probably the, you know, at some level one of the biggest pools of,
01:30:41
you know, ownership that's going to unlock. >> And then, you know, I do think a lot of
01:30:47
people probably own SpaceX through SPVS, and those probably don't unlock or get distributed anytime soon. So, I just,
01:30:53
you know, I would be careful with assuming I don't think Elon's a seller of his >> No, he's definitely not, obviously.
01:31:00
Yeah. Yeah. So, >> and a lot of us and a lot of us aren't sellers, you know, and that's great.
01:31:04
Yeah. >> Yeah. I mean a lot of people who had large SpaceX positions were large buyers
01:31:09
on the IPO on Cerebras. So Cerebras has had a a tough two days since they reported their
01:31:18
first quarter as a public company. And I think there there there are two things that are very that are worth discussing
01:31:25
here. One is there is a whole generation of portfolio managers. There's a lot of
01:31:31
people who are advocating for kind of squeezing the blood out of the stone on IPO prices. And the flip side of that is
01:31:38
that there are a lot of portfolio managers who if a stock breaks deal price, they sell it no matter what. They
01:31:47
consider it a promise that was broken. And so this is what has happened with Cerebras
01:31:55
to some degree over the last two days. And you know this may seem irrational but there are people who run giant funds
01:32:02
who I know personally where if a stock breaks deal price they sell no matter what. And so if stock breaks deal price
01:32:09
it can sometimes you know go to places you wouldn't think it would go. And this means that shorts if a stock gets close
01:32:16
to deal price they short it because they want to break deal price and then you know they make a quick 10 or 20%. So it
01:32:22
becomes a pylon >> because you have this price insensitive selling that can be triggered and this
01:32:27
is what has happened to cerebrus. You know people talk about hate sale hate selling but they broke deal price and so
01:32:34
just if you're going public and you're listening to this >> tell your bankers price this in such a
01:32:40
way that we're not going to break deal price in our first nine months as a public company. And that's what I always
01:32:46
advise everyone to do and I think it's important but I also think you know it takes companies a while to learn how to
01:32:55
tell their story and communicate to to public markets. It's a very different audience than VCs and the way like the
01:33:03
way I would have respectfully told the Cerebra story because what happened to Cerebras is they reported a quarter and
01:33:09
and they're growing fast but relative to the rest of AI they're not growing that
01:33:13
fast in the March quarter. So what I would have said is we signed this transformational
01:33:20
you know 202$25 billion I don't know the exact number contract with open AAI in December December of 2025
01:33:29
we immediately ordered more wafers from Taiwan semi takes 100 takes four months from when we make that order Taiwan semi
01:33:37
you know says yes they start producing takes four months to make the chip then we then it takes us two months
01:33:46
plus or minus to turn that chip into a server. And then if we're lucky and we can find the power, it takes us a month
01:33:54
to energize that chip and start making tokens with it. So the first time you're going to see the impact of this open AI
01:34:03
deal at the earliest is probably around Labor Day. So you'll see a little bit of
01:34:08
it in the in the third quarter, but then it really it starts to build and just like really simple math. So like let's
01:34:16
just use some rough numbers. Let's say let's take some Nvidia numbers. It takes them 35 billion to bring on a gigawatt
01:34:23
and 15 billion that you can generate 15 billion in token revenue in cloud revenue out of that gigawatt. And
01:34:30
somebody on the call talked about adding 50 megawatts a month. If they could add
01:34:34
50 megawws a month in 2027, forget 26, that means they exit the year at roughly a 9 billion cloud computing run rate.
01:34:45
And you know, we're at we're at less than 40 billion of market cap. Now, that is going to be really hard to do and
01:34:53
they've never done anything like that before. And but I would I would I would focus like as an investor, I think what
01:35:01
matters here is not where they sit competitively, not what new demand they can bring on,
01:35:07
but just how quickly can they bring on power. And listen, like outside of the hyperscalers, the only companies that
01:35:13
have ever brought on more than a gigawatt, I think, are Cororeweave, um, Crusoe,
01:35:18
um, and, and, and, and, and SpaceX AI to bringing on 600 megawws, it's really hard. And that is what like I'm focused
01:35:27
on as an investor. How many megawatts can they bring on? Because we know what they're going to monetize at. And that
01:35:34
is the question. And we'll see. >> Yeah. And I think that's well said. >> Yeah. the other companies that have had
01:35:41
this happen of late. I guess Rivian and famously Chimath Facebook traded below IPO price in the first year I think. And
01:35:48
so this isn't necessarily mean a bad company. It just means a lot of hype or maybe going for it in the IPO and
01:35:55
pricing it to perfection. And it can go one of two ways. You can either grossly underpric, grossly overpric. And it's
01:36:02
really hard to get right. And this is why doing auctions >> or you know >> Yeah, auctions are the way. Auctions are
01:36:09
the way. And so it's not hard to price if you just do it the way it's supposed to be done. It's hard to price when
01:36:14
you're when you uh want it to be a certain number. >> Yeah. And you have many mouths to feed
01:36:21
and that influences the price. >> I don't think that happened here. I think there was a good faith effort to
01:36:27
to price this thoughtfully, but just, you know, there was it was such an in demand IPO. I think it was a hard IPO to
01:36:34
price. You just auction it, Gavin. You just auction it. I think that's different
01:36:39
than underwriting it. You're underwriting, >> but the bankers like I think they just
01:36:44
need to get more in the mode of just doing the auction, you know. >> Yeah. Final topic was this uh $3
01:36:52
trillion or actually now it's your $3 trillion call on anthropic would make it like 6
01:37:01
trillion in offerings. What do we think broadly? in offerings because it's they're going to offer a small slice of
01:37:07
that. >> Sure. >> And that small slice is very e easy to absorb in the context. I believe we have
01:37:13
no precedent. We'll see. I might be wrong, but it's not hard for global capital markets to absorb
01:37:19
>> 50 billion of an offering. It's not like it's not like somebody has to come up.
01:37:24
>> Yeah. It's a 5% float is 150 billion. There's enough >> guys. Remember remember when 15 billion
01:37:30
used to be a whole lot of money? >> Yeah. How weird is what's going on? How weird?
01:37:35
>> By the way, crazy. Where is all this money coming from? Oh yeah. Where was it
01:37:39
all this time? >> Can we just reminisce, man? >> We got my when I was at Fidelity, my my
01:37:45
colleagues and I, we got pillaried. >> I think we priced Uber at 14 billion. And we got pillaried in the press for
01:37:53
not knowing what we were doing. And it was seen as, you know, then then of course like six months later, I think
01:37:58
you guys did around at what, like 42 billion or something. But it's like, yeah, 14 billion. That was
01:38:04
groundbreaking for a private company back then. >> Yeah, for sure. It was 17, but that, you
01:38:08
know, no 17. No, you're right. I wanted 14 >> and we had like a big negotiation >> and we ran we ran an auction. No, we
01:38:17
just ran an auction. >> Yeah, you wanted 20 and maybe we landed at 17. What we did, what we did was
01:38:24
every person who wanted to be involved had to fill out a sheet of how much money they put it at 10, 11, 12, 13, 14
01:38:32
or all the way up to 20. And then we just did the Dutch auction. We said, "We want to clear one and a half bill." We
01:38:38
just did the auction and cleared it. Went back to people and said, "Hey, you're not going to get it.
01:38:44
You have another shot." they update their Excel sheet and it just moves the number up a little bit and then you you
01:38:51
you close it down. But yeah, it started that round started at I think it was 9 or 10 or something like that. Ended up
01:38:58
at 17. >> Thank you for your service. >> Yes, it used to be it used to be that that was a lot of money and that was
01:39:06
unheard of at the time and that was well look it was 10 years ago when we used to
01:39:10
walk to school uphill both ways. >> Yes. We used to have to Yeah. We used to eat potatoes with no butter, just hot
01:39:18
potatoes. Sometimes we were lucky that they were cooked. Sometimes they were raw. No, my dad would tell me that
01:39:23
story. It's a famous John the Beard story. He'd say his mom would put like four um four potatoes in the oven.
01:39:31
They'd wrap them in tin foil, one in each pocket because they couldn't afford gloves. So, you put your hands in your
01:39:36
pocket with the hot potatoes. You get to school, you eat one for breakfast, one for lunch. him and his sister uh
01:39:40
Johanna, God rest for so my aunt who died too young and they would just go eat these potatoes. That was their life.
01:39:47
Uh walking to school. >> Yeah. And our kids are trying both ways. >> Yeah. They're trying to get the new
01:39:54
iPhone 16 or 17. I don't know. >> Remember when Remember when it used to be hard to raise $5 billion?
01:39:59
>> Oh my lord. This is crazy. >> Remember when it was hard to raise$ 1.5 million?
01:40:04
>> Dude, I remember Yeah. I remember when it was hard to raise. Anyways, we don't
01:40:08
have to do this. It's okay. We already sound like grandpas. It's fine. >> We do sound like It was so hard to raise
01:40:14
that first 1.5. >> Dude, it's just We don't need to go there. It's so funny though.
01:40:19
>> What's with this art piece, Gavin, behind you? Is that like an art piece or you made it?
01:40:24
>> Yeah. >> No, this is I'm staying at a rented house in Atheertton. >> Oh, okay. There you go. It's just a
01:40:29
Arabian art. All right, everybody. for the dictator Jamal Pitia and for David Saxs TK GB.
01:40:39
We'll see you next time on the Allin podcast. Bye-bye. Great job everybody. >> We'll let your winners ride.
01:40:47
>> Rainman David. >> We open sourced it to the fans and they've just gone crazy with it.
01:40:55
>> Love you. I want your winners. >> Besties are gone. >> That is my dog taking your driveways.
01:41:12
>> Oh man, my appetiter will meet up. >> We should all just get a room and just have one big huge orgy cuz they're all
01:41:18
just useless. It's like this like sexual tension that we just need to release somehow.
01:41:26
>> Your feet. We need to get merch. Going all in. I'm going all in.