SpaceX’ $75B+ Historic IPO, GPT 5.5 Outperforms Polymarket, and AI Solves 80 yr old math problem | EP #257

SpaceX filed for what's expected to be the largest IPO ever. $75 billion being raised at a valuation probably north of $1.75 trillion. He's about to have a currency to go on a shopping spree.

That sentence is so out of band with any point in human history. So anyone out there doing a startup, there is capacity for 1,000 unicorn transactions. Remember, unicorn is called unicorn because it's supposed to be extremely rare.

GPT-5.5 is now beating prediction markets on forecasting the future. It beat Polymarket crowd predictions for the Super Bowl.

This is the worst psychohistory models will ever be.

The concentration of wealth effects from this would be insane.

The financial singularity.

This is a problem posed by the famous Hungarian mathematician Paul Erdős about 80 years ago. Now AI has just solved it. Not only was it faster, not only was it able to brute force, but it was also smarter.

This is like a much bigger moment in history than just solving a math problem.

#SolveEverything, Peter. We're seeing it.

Now that's a moonshot, ladies and gentlemen.

Welcome everybody to another episode of Moonshots. I'm here with my extraordinary moonshot mates, Saleem, the father of organizational singularities. You're spawning singularities everywhere, buddy. Alex Wiesner-Gross, our in-house polymath. Dave Blunden, our wizard of AI investing. I'm Peter Diamandis, your host, and hopefully your abundance whisperer for an optimistic future. You know, we've loaded the show today with extraordinary stories, uh, hopefully stories that get you excited about being a builder, literally starting to build before this episode is out. Gentlemen, good to see you all. I have to ask, uh, our normal where's Waldo question. So today, uh, Dave, where are you, buddy?

Uh, back at Stanford, had a whole bunch of really fun meetings with AI founders here.

Awesome. And Salim, I, I know you're not home because you're never home. Where are you?

No, I'm in Brazil like we did yesterday, so I'm still here till tomorrow and then I fly back.

Okay, well, and Alex, you and I are in our normal haunts. Um, that's right. Yeah, uh, amazing. So, uh, let's begin. You know, we've stacked the show today. Here's a quick look and preview of what we're going to be covering. SpaceX just filed their biggest IPO in human history. Uh, it's extraordinary. And later today, after we record this episode, we've got Starship V3 scheduled to launch. OpenAI, uh, has just disproved a conjecture in the mathematics realm. We'll be talking to Alex about what that all means. GPT-5.5 is now beating prediction markets on forecasting the future as well. ChatGPT just became your financial advisor. A lot to cover. You know, our mission: keep you optimistic, informed, and ready for the supersonic tsunami heading our way. So with that, uh, let's get started. Our first story here is in fact the SpaceX IPO. SpaceX filed for what's expected to be the largest IPO ever, $75 billion being raised at a valuation probably north of $1.75 trillion. Uh, the biggest in history, you know, over 2.5 times that of Saudi Aramco. Uh, Elon is maintaining his super voting rights with, uh, insiders controlling 86% of the voting power. Uh, and I love this— SpaceX's IPO prospectus says it expects an addressable market of $28.5 trillion. I mean, that's quite a TAM. Dave, let's go to you first on this one.

You know, the TAM is just under the size of the entire US GDP, and I'm sure they landed it. You know, we don't want to claim to be bigger than the entire US, so we'll be just one notch below. But Erik Brynjolfsson hosted Maura and I for dinner last night at his house. Erik is a professor of economics and AI at Stanford HAI. That's the AI lab of Stanford. And he was the first guy to mention this to me. And I was like, come on, $25, $28.5 trillion TAM. So he was going off on, uh, yeah, but, um, there's no way to disprove it. Uh, there's no reason it shouldn't be true. You know, Elon, remember Elon's core thesis is that we can 10x the global economy in 10 years. Most people think it's possible, but it's longer than 10 years. But regardless, if the, if the global economy 10x's, then his TAM should easily fit within $25 or $28.5 trillion.

I looked at the breakdown of how he got to $28.5 trillion, and it's interesting. So $870 billion, you know, just a mere bit under a trillion is Starlink's business. $740 billion is Starlink's mobile unit. $600 billion is their digital advertising market through X. $2.4 trillion is their AI infrastructure. And get this, $22.7 trillion is kind of come from MacroHard. If you remember, MacroHard is their partnership with Tesla where they want to emulate all digital work and create an AI-run software company. So pretty spiffy. I mean, a trillion here, a trillion there adds up to a lot, a lot of money. Alex, what do you think of it?

You almost have to ask, did SpaceX acquire xAI or did xAI acquire or reverse acquire SpaceX? Certainly based on the TAM analysis, it looks a little bit more like the latter rather than the former with enterprise applications dominating the addressable market. I think the most interesting part in the entire prospectus was actually the bit about what Anthropic is now paying SpaceX, $15 billion per year for data center access.

Yeah.

And by the way, in the past 48 hours, it's not just Colossus 1 that Anthropic is paying for, it's Colossus 2 as well. So in the past couple of pods, I made remarks to the effect that Groq is on life support. And a bunch of folks I saw in the comments said, oh no, he's, he's against Elon. How outrageous is this claim? No, actually, I think far from being outrageous, I think SpaceX's prospectus arguably supports that entire thesis that SpaceX is basically taking now not just Colossus 1, which maybe charitably one could have argued is sort of a relic older, slightly older GPUs and heterogeneous set of GPUs at that, and giving it to Anthropic. But Anthropic now also using compute from Colossus 2? No, this to my eye looks like SpaceX basically abandoning the foundation model space, handing it over to Anthropic, focusing on the infra layer, building out the Dyson Swarm. Record time to mention Dyson Swarm. Record time to mention Dyson Swarm into the episode.

Well, just taking—

and focusing on becoming the infra layer. And they can— I think SpaceX, I'm just reading the tea leaves here with the prospectus. SpaceX AI seems relatively uninterested at this point in owning their own foundation model. They've also, as part of this, announced that they are going through with the purchase of Cursor 30 days after their planned IPO. And Cursor is based on KIMI, at this point. So SpaceX's foundation model lineage is seemingly switching over to a derivative of Chinese open weight models with a lot of fine-tuning based on probably American reasoning traces. But SpaceX in this weird climate wants to seemingly own the layer below and above. It wants to, to own the infra layer. That's the Dyson swarm and all the data centers. And it wants to own the layer above. That's macro hard. Which other hyperscaler does that start to look like? Microsoft with its strategy of— Microsoft OpenAI.

Exactly. Are they still playing?

Meet the old boss, same as the new boss. SpaceX is trying to transform itself, I think, into a Dyson Swarm version of Microsoft.

I was on CNBC this morning doing an episode about the SpaceX IPO, and one of the things I was sort of like just shouting from the hilltops is, listen, you can't think about SpaceX just from their Starlink revenue or even their launch revenue. What SpaceX is doing is opening up the space frontier and everything we hold of value, you know, metals, minerals, energy, real estate is in near infinite quantities in space. This is not just, you know, the first ships from Europe to the US. It's the galleons, it's the railroad and SpaceX. As they open up this transportation infrastructure, they're going to own the businesses all along the route. And at the end, what we're building there on the moon, in Earth orbit, in the inner solar system and beyond, you know, if you look at this, Alex, I know we've talked about this before, uh, Starship, uh, compared to all the other launch vehicles, right? We can compare it to New Glenn or to Relativity Space, what they're building, or what Rocket Labs is building with Neptune. Doesn't come close. I mean, Starship is planning to launch on the average of once an hour. Today, you know, Falcon 9 is launching every 2.5 days. And when they get to an hourly launch rate, airline operations, they open up the most extraordinary wealth in the universe.

But the interesting thing to me is that they're, they're barely even selling it. Like, yes, it is in the prospectus that they intend to provide Earth-to-Earth transport to humans, cargo transport type things. Obviously, they're still a space heavy lift company. But if you just look at the prospectus, they look like Microsoft in space. That's the story that they're selling to retail. Right now.

Because that's what the investors understand.

That's legible to capital markets.

I think it would freak people out. I think it would freak people out. I go down the same path you did, Peter. This is planetary infrastructure. It's like Christopher Columbus sailing off to kind of trying to colonize a whole new world here.

Yeah.

Well, just some confirmation for what Alex was saying too. The Stanford PhD I was just meeting with before this podcast, tracking all of the talent and confirmed that, yeah, the great people have left xAI by the droves, actually, but they're all going to Anthropic.

Yeah.

And, you know, Karpathy, I mean, he's got to be the ultimate coup. We had that on the last pod. Shane Longprey from MIT, who's phenomenal, he's joining Anthropic now too. So a duopoly between Anthropic and SpaceX AI. Is pretty daunting for everybody else. You think about the best researchers all want to join Dario because they trust him and because the models are phenomenal. And then you put all that compute onto Elon's space empire and that's a heck of a duopoly. I don't know though if Elon plays duopoly forever and it doesn't seem like his—

His playbook.

Yeah.

He doesn't play well with others. It's fascinating.

So, you know, the other thing, the TerraFab, you know, where's the TerraFab in the prospectus? It's maybe it's the same story. We don't need to, you know, promote things that are still hypothetical. This is enough. $1.7 trillion is enough for now, maybe.

Yeah, there were a bunch of assets that were shared. MacroHard itself was, if I remember correctly, characterized as in part shared with Tesla, in part being informed by Tesla AI. And I also think Optimus, there's There's a curious split that I don't know how it's going to play out between the Optimus AI and the macro hard/XAI cursor AI, where right now SpaceX in this bizarre sort of incestuous ecosystem of Elon-affiliated companies, SpaceX is seeming to get in some sense the digital Optimus, if you will, the macro hard worker for knowledge work labor, and Tesla is seeming to get the embodied Optimus. And not obvious to me exactly how from a governance perspective all of this IP is supposed to flow back and forth between them.

A lot of predictions, Alex, about merging obviously SpaceX, AI, and Tesla into, you know, Musk Corp. I don't— I wonder if there's a poly market on the under-over there. But, you know, a lot of it—

there is a prediction market for it.

Yeah. Do you remember what it might be? Maybe you can, maybe you can check in the background.

Yeah, I'll check.

But honestly, I think within a year we're going to see that, you know, when we've got two publicly traded markets, the ability to value them and merge them becomes a lot easier.

Well, the bullet 2 is really important in that, the super-voting control. Remember we mentioned on a podcast a while ago that Elon has never had what Sergey and Mark Zuckerberg have, which is a public entity that can raise, you know, many billions of dollars in an overnight. Where he's the controlling shareholder. So this will be the first time that he's in that position. Remember, he doesn't control Tesla. That's why he's constantly going to Delaware court over his compact. Yeah, this'll be, this'll be a new thing in the Elonverse. And remember, Peter, you know, you were an early investor in XAI. Remember that capital raise when it was getting off the ground?

Yeah. The very first one. Yeah.

Remember the amount that he was raising, which seemed like a lot at the time?

Oh God, I don't. What?

I recall— I don't remember either, but I recall it being like an $8 billion valuation and maybe a billion-dollar raise or something like that, which is just hilarious now because you're, you know, now you're looking at a $100 billion, you know, size, the $75 billion here raise.

He's about to have a currency to go on a shopping spree.

Yeah.

And so my bet is once they go public, they're going to be beginning to acquire a number of companies as part of it.

And one thing we posted the other day, uh, once these guys are public, they can easily do 1,000 acquisitions or more of $1 billion or more. The world— like, that sentence is so out of band with any point in human history. So anyone out there doing a startup, you know, yeah, maybe you'll screw it up, but assuming you don't screw it up, there is capacity for 1,000 unicorn transactions. Remember, unicorn is called unicorn because it's supposed to be extremely rare. All right. It's just a whole new world.

The old days.

And by the way, the prediction market, uh, I have it for you. Polymarket predicts by the end of this year, a 20% probability of SpaceX and Tesla merging.

Okay. All right. Uh, end of this year, but I think within a year, I think that's going to happen. Things settle out.

Yeah, that's right.

Yeah.

All right. Uh, you know, I wish we were recording this later. I'm so excited. Uh, what's coming up today is the launch of the Block V3. Starship, 100-ton payload capability to orbit. It's an extraordinary vehicle. You know, we're talking about a thrust of 18 million pounds using the Raptor 3 engine, the most beautiful, elegant engine ever. This is Flight 12. On this flight, they're going to be demonstrating the docking ports that are going to enable orbital refueling. Of course, orbital refueling of Starship is required for the lunar missions they're planning, uh, to win, as well as going to Mars. Both stages are going to be splashing down, uh, the Super Heavy in the Gulf of Mexico and Starship in the Indian Ocean, as it's done before. And hopefully they'll have buoys out there to watch both of them. Uh, remember the NASA Artemis mission. Artemis 3 is a docking test later this, uh, in 2027. Artemis 4 to the lunar surface, particularly the South Pole, is taking place in 2028. Alex, uh, you'll be watching, of course.

I'll be watching, not from the moon. I, I would really love to see orbital refueling happen sooner rather than later. That's— I, I don't know when that's expected, later this year, but I, I think being able to get to the point where we cannot just do propulsive landing but also orbital docking and refueling between Starships is going to be such a key moment. It's also the, the hammer that the Bezos's of the world have been using to argue that the Starship architecture, which is a little bit— if I were to play out an analogy, a little bit more internet packet-oriented. It's the, the lunar architecture of Starship colonization of the Moon is very launch-intensive. It involves dozens of independent launches and refueling steps, whereas historically, if you look at like the Apollo architecture, was a much more monolithic architecture, you go up and then you go over and then you land and then you come back.

You carry all the fuel from the ground for that one mission all the way.

Exactly. So I think the world, myself included, will be watching with quite a bit of optimism that, call it the Starship packet type architecture for you send everything up to LEO in packets and then you do a bunch of refueling steps and then you send the packets over to the moon or you send them to Mars is a superior solution. I would view it almost as analogous to the switch, no pun intended, that the internet made, or I should say that networks made from circuit-switched systems where you had in some sense a single contiguous bit of atoms connecting a transmitter and a receiver on a network to packet-based switching where there was a complete decoupling of bits from atoms. Same idea here with a Starship-based solar system transport architecture. We're aggressively decoupling cargo from transport. And so fingers crossed that, that Starship V3 and orbital refueling later this year enable us to packet switch the solar system.

You know, I don't think people realize how unique Starship is in terms of how it was designed, right? This is designed for full reusability. It's designed to land, refuel, and go again. It's, you know, his vision is airline-like operations. And the amount of throw weight outstrips everything else. I have a, you know, the one thing that Elon has done extraordinarily well is his manufacturing approaches and his sort of building for an ultimate capability. You know, New Glenn, again, Relativity Space now CEO, friend of Pod, Eric Schmidt, that those vehicles I don't think they can compete. They're going to have to build new capabilities to get to this launch frequency that enables all of these visions.

Well, if data centers— you heard Sundar at Google I/O saying data centers in space, we're not going to talk about it today, but it's very clearly on the roadmap. But if Elon and Anthropic are working in cahoots to, you know, to to build the Dyson swarm or Dyson sphere, then Google has to react to that with something. And here's Eric Schmidt, our former CEO, with a rocket company. So they have to play fast followers somehow. I mean, you're right, it's starting from pretty far behind, but something has to give because Google obviously can't buy SpaceX.

I also wouldn't necessarily over-index on the Starship architecture being the definitive final word on how we get to orbit. There are going to be many, many future technologies I would predict over the next 5 to 10 years that will leapfrog in terms of the applied physics Starship's launch capabilities. And it won't necessarily be SpaceX that's the leader in the field for leapfrog capabilities. So I think there are many ways to orbit.

The development time takes years, right? We haven't seen a vehicle go from, from design to operational flight in anything. I mean, even, even Falcon 9, uh, to get it to reusability, get it to the, the rate of success, right, 99.99%, if you would, uh, takes a good 5-year time. And in the interim, I mean, don't forget, Elon's going to be designing whatever follows Starship as well.

That's true, but there are two things happening concurrent with that. First of all, the fast follower effect. You know, once you can see what worked and what didn't work it really helps you a lot in the copying. And that varies by technology. But then concurrent with that, you have like Mikado coming up with mechanical design AI. You know, you remember when we were interviewing Elon, he said, this is the greatest thing ever built by humanity without AI assistance. It'll be the last great thing, but it was entirely built with people and protractors and crayons. Like, I'm kidding. But it was catchy.

How retro.

It was CAD/CAM, you know, hand-built, designed CAD/CAM without AI mechanical design assistance. That'll never happen again. So that might really accelerate Erickson.

So by the time I watch it, the design space of aerospace lift and heavy lift is vast and we've only scratched the surface of it, I think. And to Dave's point, once it's demonstrated how large the market for heavy lift is in the form of building out that Dyson swarm, I expect many, many competitors to come out of the woodwork, many of which are probably already known names, and some of them, or maybe all of them in aggregate, I do think will give SpaceX a run for its money.

All right.

We could put a bet on the side on that. I think they will ultimately, but I don't think that's going to happen between now and, you know, 2029. And then you're part of NASA's infrastructure and then you've built the original, the initial Dyson Swarm.

Also, don't sleep on China. China is busy copying everything it can out of SpaceX, and I'm sure there will be Chinese heavy launch capabilities and multiple Chinese Dyson swarms as well.

They've, they've tried. No successes yet. No successes yet. All right. Let's move ourselves out of this. And by the way, by the time people have watched this episode, you know, Schrödinger's cat has happened. Either, either V3 of Starship has succeeded or it's failed. And if it's failed, hopefully there'll be another one following quickly. This has been 5 months since the last launch. Uh, hopefully now that V3 is up and operational, we'll see a lot more frequent launches. And the one thing you got to appreciate about Elon is he's not a fa— he's not afraid to fail forward. All right. Uh, moving on. Uh, I love this article. Uh, this story, GPT-5.5 Codex is leading at forecasting. So OpenAI built something called FutureSim that replays the internet a day at a time, day by day. It gives AI agents access to the real news starting from January 1st, 2026 onward, then asks them to forecast real-world events over the next 90 days. So GPT-5.5, uh, is running Codex and scoring 25% accuracy, leading across all the frontier models. Uh, it beat Polly Market crowd predictions for the Super Bowl. Uh, and you know, the way I think about this, this is the beginning of giving AI wisdom. You know, I've said this before, if you think about what wi— human wisdom is, is the ability to go to the elder council and say, okay, you know, which direction do I take? Which way do I go? And the elders say, If you go this way, based on our experience, it's not going to end well. You go this way and it's likely to win. And if AI is able to run high-resolution simulations a billion-fold, you know, a billion times, it's going to say this is the highest probability of success. Alex, what do you think about that?

Remember Isaac Asimov's Psychohistory from the Foundation novels?

Yes.

Right. So, so the premise of the Foundation novels is that a mathematician named Harry Seldon. This is grand galactic scale sci-fi. Harry Seldon, a great mathematician, invents a theory that he calls psychohistory that's able to predict at grand scale human civilization. And he's able to predict the collapse of the Galactic Empire and a whole bunch of other interesting things. So FutureSim is a benchmark. Just a minor correction, I think FutureSim itself as a benchmark is not from OpenAI. It's from a group of independent researchers. It does benchmark a bunch of models, including models from OpenAI. And it's such a clever architecture. It's benchmarking the ability for a variety of models to predict events beyond their knowledge cutoff date and without access to the web. So they can't look up, say, more recent information. And the fact that the state of the art right now is already at 25% accuracy, this is the worst psychohistory models, if I may borrow from Asimov, will ever be. And I agree with the contention that if you extrapolate this out, we'll get Monte Carlo tree search of policy decisions. We'll be able to predict to some extent predictable events in the future, at least as a function of perhaps human actions. And Peter, you and I spoke about this a little bit in Solve Everything in some of our forecasts for what would happen the latter half of the decade from 2026 to 2035 about planetary-scale solutions. I do think with the ability to predict planetary-scale outcomes come the ability to predict planetary-scale interventions, not unlike how I would argue disease is probably going to get solved. This will be the planetary analog of what a virtual cell can do for curing all disease in the sense that if you have a perfect digital twin of the system that you're trying to fix, you can exhaustively test all possible interventions to get you from the bad state to a good state.

One of the greatest tools we could have for a positive outcome for humanity. Celine, what do you think about this?

Well, this is going to be incredibly powerful for the boardroom because you go from quarterly kind of updates to real-time sensing. We've actually built this into the architecture following your paper into the organizational singularity architecture, and I'll talk about that a bit later.

There's another implication of this that we're kind of glossing over, which is, you know, right now if you look at what New York does, there are many, many, many hedge funds that specialize in different areas. So, conductors and retail and whatever. And those are all supported by prime brokerages. So there's much bigger banks that do all the trading and accounting, and, and those are prime brokerages. That entire industry could turn into just one or two AI models. Um, and so the concentration of wealth effects from this would be insane. But if the AI is just fundamentally better at picking markets, it's not going to sit there and just do one market. It's going to expand quickly across all markets. And so you're going to, if this plays out the way it's starting to, you're going to see a collapse into just a couple of mega funds that have massive AI budgets.

The financial singularity.

Yeah.

Did I just hear Dave and Peter, did I hear you argue in favor of indexing versus individual stock picking?

Oh God, no. That's not an index. The index finds— this is so much better than an index. Well, actually it's an active index. I'll call it a— A half agreement, if you call it an active index. Okay.

Well, you know, the financial markets are still, you know, Dave Erickson, you and I have been going back and forth on our, on our texting back and forth. Just, you know, the predictions that Leopold made on energy infrastructure and even reducing his level of interest in chips are still playing out. You know, we've seen this, the demand for tokens is outstripping supply, and what's keeping that limited is energy and infrastructure. Hey everybody, you may not know this, but I've got an incredible research team. And every week, myself, my research team study the meta trends that are impacting the world. Topics like computation, sensors, networks, AI, robotics, 3D printing, synthetic biology. And these meta trend reports I put out once a week enable you to see the future 10 years ahead of anybody else. If you'd like to get access to the Meta Trends newsletter every week, go to diamandis.com/metatrends. That's diamandis.com/metatrends. All right. Uh, let's stay with OpenAI. So OpenAI launches personal finance in ChatGPT. Uh, this is, uh, a finance mode that is able to access 12,000 financial institutions, uh, letting, uh, ChatGPT Pro users ask personalized questions about their spending, their debt, their taxes, long-term planning. This is OpenAI eating another vertical. Uh, they did it to search, they're doing it to coding, and now they're coming after a $12 billion personal finance app market. You know, one could say maybe Mint will be dead. NerdWallet, you know, should be nervous. 200 million people already use AI for financial questions. Dave, what do you make of this? And Celine, let's go to you next after Dave.

Well, this is part of an overall trend where the foundation model companies are starting to roll out legal. Now they're rolling out finance, they're rolling out APIs. That enable vertical disintermediation of all these vertical companies, lawyers, financial accountants, whatever. I think it's creating an ecosystem of new startups that are early adopters of these APIs that can then disrupt the markets. And I think those companies are going to do incredibly well while the foundation model companies do incredibly well. But what a lot of people are overlooking, if you're in New York and you look at this massive financial institution with 20,000 financial advisors, You can't envision it getting disrupted. There's just so much mass and concrete and meetings and files and so much regulatory barrier. But over here, this parallel economy is growing, which is the AI economy working within itself. And everyone in New York is like, yeah, but all the money's over here in the banks. It's not over there in that new economy. Well, after these next IPOs, that money will be in the new economy. It goes through the public markets. Through SpaceX and through Anthropic and through OpenAI back into the agent-to-agent economy, which is an entirely parallel banking and finance system independent of the original one.

Which will outstrip everything else.

Yeah. Well, if Elon's right, it'll be 10 times bigger than everything you see in New York in about 10 to 20 years and growing on a much faster curve. And it doesn't care about any of your legacy baggage. It's going to grow on its own. And so I think a lot of people are coming around to this view that we can leave the concrete world alone. We don't need to scare everyone and disrupt it because we're building a parallel AI world here anyway, and it's going to be bigger anyway. And so I think these— it's an interesting story within the story. Love, Celine, I'd love to get your take on this.

Yeah, so this is essentially the solve everything thesis playing writ large, right? You move from traditional models to an inner loop of intelligence and everything else wrapped around that. That's exactly the architecture we have in this organizational singularity stuff effort on, the banks should be terrified because the interface to money is shifting away from them to the AI, right? And they're going to lose control in a very dramatic way. The, you know, financial device will get layered around this, not around an individual. And I think the other broader point is that you made, Dave, is that we're moving. We don't have to disrupt the legacy. You build this completely new architecture and just let that become the new gravity center. Go straight to the Buckminster Fuller old quote that he said, you can't fix an existing system. You have to set up a new system at the edge and let that become the new gravity center. We're going to see that happen in legal, in healthcare, in insurance, in education. It's just going to happen over and over again.

I think the thing that filled in, just filled in this month and is so newsworthy is the conduit of the money from the old to the new is now really clear. Like Orne, you know, where Alex is an advisor, is a really good example. Orne is a security. Where you can deploy your money into AI and you can get it right on Robinhood or buy it on the exchange. And that conduit then moves the money from the old economy into the new economy. And once it's in the new economy, it doesn't care about the legacy banking system. And the same with these IPOs that are coming up, trillions of dollars that are moving.

It's like, it's like the Bitcoin ETFs. Once you've got it over there, you don't care.

Yeah. Breaking news, by the way, on this, and I'll hand it to you next, Alex, but breaking news is that OpenAI Uh, is, uh, sort of letting it be known they are preparing to file for their IPO as early as this week, as this Friday. You know, they just won the case against Elon, uh, they're feeling their oats. And we talked about this a few pods ago, it's a race to get access to the available capital. SpaceX is going to suck a lot of the oxygen out of the room. Uh, you know, their competition Anthropic is preparing for an IPO. I mean, it's interesting Remember, we had had conversations from the CFO of OpenAI, Sarah, saying they were going to file in 2027 because they weren't ready. But here they are talking about filing, you know, ahead of Anthropic to get access to that cash flow. God knows they need the capital to build out their compute. Alex, over to you.

Well, as I closed my newsletter today, the day of recording, Cogito ergo IPO. I think therefore IPO. Is the thinking of the moment. When I look at the announcement of OpenAI launching personal finance, I ask myself, where's the monetization? OpenAI having, as I've mentioned numerous times on the pod, having largely pivoted away from consumer to enterprise and needing to justify very high value productivity per token, I ask myself, where's the value per token in this? Is OpenAI really seeking to become a financial institution Doubt it. The value, I suspect, is a bit of Kremlinology, is advertising. OpenAI is following the Google playbook. Why did Google go into developing all of these financial verticals? Because some of the financial queries, and Dave probably knows better than any of us, the financial queries can be enormously lucrative from an advertising perspective. So if OpenAI hopes to monetize personal finance conversations, I think they're probably just going to do it by running OpenAI ads ads for consumers relating to personal finance. And the more that they can tailor particular information to the particular circumstances of a retail investor with all of these integrations, that will enable them to target far better ads in conversation to those users.

And today Dario announced no ads on Anthropic, just flat out, we're not doing advertising.

Which is very convenient post hoc given that they're focused on enterprise.

Yeah, they don't have the consumer anyway.

So it's very easy to be an angel investor.

Yeah, great ethics.

Yeah.

Yeah, there you go.

By the way, this one more thing about this was this feels also like a bit of a copycat after Anthropic launched all these plugins for legal, et cetera. Here comes OpenAI doing the same thing for personal finance.

But critically, Anthropic's suites of skills are targeted at businesses, whereas OpenAI is targeting this at consumers. OpenAI wants to charge enterprises. So I'm sticking with my ad theory.

Yeah, that's right. Google too. It's OpenAI versus Google watching this ad. Evolution.

I think you're right. Yes, I think you're right, buddy. Okay, uh, I'm gonna turn this over to Alex. Here's the story: OpenAI model disproves a central conjecture in discrete geometry. So an OpenAI model today, or this week, disproved a long-standing conjecture from Paul Erdős, one of the most prolific mathematicians in history. Alex, tell us about it. Uh, actually, let me run this video and then give us your blow-by-blow. What does this mean? What does this mean to the average listener?

I think what's significant about this moment is that it's the first really clear example of AI solving not just an unsolved maths problem, but a really well-known unsolved maths problem.

This is the first mathematical breakthrough due to an AI.

It's been described as the most well-known problem in combinatorial geometry.

So for, for a whole subfield of mathematics, it's like maybe the best-known problem there is.

So I remember seeing an initial version of the model output.

I sort of didn't really believe it. It took quite a while sort of reading over, trying to figure out— this problem is about points in a plane. It's a completely elementary geometric problem, but the solution involves really deep tools from algebraic number theory.

And it was believed that the construction was, was basically best possible. But what our model did was show that this construction could actually be improved by quite a bit.

Alex, your explanation, please.

So first, just a few seconds about what the problem is. This is a problem posed by the famous Hungarian mathematician Paul Erdős about 80 years ago. And the problem was basically asking the question, if you have a plane, a two-dimensional plane, and you can put n points in the plane, what is the maximum number of pairs of points that can be separated by the so-called unit distance, basically by a fixed distance? It's a very simple to pose problem, very hard to answer. So Erdős's original conjecture, which has held essentially until now, was that it was effectively impossible to do much better than in terms of the number of pairs that could be separated by this fixed unit distance. It was, he conjectured, impossible to do materially better than some number that's proportional to the number of points themselves. So basically some, some number of pairs that's linear in the number of points. And now for the first time, OpenAI has revealed that an internal model that hasn't been publicly released has disproven that conjecture and found weakly superlinear scaling. Why should anyone care? This is just, as I like to say, math is cooked. This is going to be the new Exhibit A that I cite for how cooked math is. This is one of the most important problems, as the video mentioned, in combinatoric geometry that stood for the past 80 years. And now AI has just solved it. And notably, if you unpack all of the accompanying documentation and commentary. There's quite a bit of interesting commentary. This wasn't, say, something like the four-color problem that one might imagine given that it involves combinatorics. Four-color problem being if you have a two-dimensional map, what's the minimum number of colors that you can tile each or color each country in to make sure that no two adjacent countries are the same color? Problems in math, in combinatorics, like the four-coloring problem, that tend to be exhaustively solved by AI. And then mathematicians and others point to those exhaustive brute-force solutions and say, gosh, maybe AI is more exhaustive, it's better at brute force, but it lacks human brilliance. It lacks leaps of creative insight. This is not a problem like that. This is a problem that has the top mathematicians in the world who specialize in this particular area looking at its reasoning traces and concluding that not only was it faster in some sense, not only was it able to brute force lots of different theoretical approaches, but it was also smarter. And it was smarter, though, in an interesting way. One of the commentaries— and definitely I would encourage everyone to go to the OpenAI website and read a number of professional mathematician commentaries on the reasoning chain of thought that it used to solve, really to disprove this conjecture. There were some interesting comments to the effect that from looking at the reasoning chain, they could see that it was pursuing all sorts of, call them exotic possibilities that humans would be too exhausted to pursue. So it was arriving at creativity by sort of both being faster, but also being able to brute brute force all sorts of outlandish possibilities. And in the end, one of those possibilities, and I think the language from the chain of thought that ultimately led to the solution was— began with something like, optimistically, if I pursued this, something might happen. And that turned out to be the solution. So if we remember the infamous move 37 from AlphaGo's match with Lee Sedol and how being able to brute force, but with clever learned policy search, the reasoning tree of a Go game, We're starting to see that play out in math. And by the way, that's going to play out everywhere else as well. It's going to play out in physics. It's going to play out in every science, engineering, starting— #SolveEverything, Peter. We're seeing it.

The starting gun. And, you know, while people may not relate to math, they sure are going to relate to physics and chemistry and biology, material sciences.

Yes.

These are going to give birth to trillion-dollar outcomes.

I think, you know, we, in the past, we've had very hard Humanities Last Exam questions on the pod and, and made the point that you would have to think hard for 3 or 4 hours to even understand the question. And then, of course, then Alex says, I think the answer is 4, and it turns out to be right. But putting that aside, this one, you know, if you rewind the video and listen to the first couple sentences of what Alex said, that's the, that's the whole problem. If you like crossword puzzles or you like Sudoku, you're going to love this one. Listen to it again. I'd really encourage the team to splice in the images because for the last 80 years, humanity thought the best solution was this really simple grid. It's just points in a square. And if you look at the final AI solution, which is not proven to be the best solution, it's just better than the square, it's beautiful and it's elegant and it's not intuitive at all. And it gives you a lot of new insight into, wow, AI is going to be really good at things like magnetic bottles and protein folding and chip lay. It looks a lot like a chip design problem. We're laying out the wires in an optimal way. The final answer doesn't look intuitive to you at all as a human, but it's better. And so this is like a much bigger moment in history than just solving a math problem. It's so noteworthy. I thought, Alex, I thought you described it so beautifully and eloquently, hopefully.

I think you did a great job too. I think that the solutions, the optimal solutions to things aren't necessarily as human legible as the human solutions. And I agree with the sentiment, Dave. I think this is heralding an era when AI solves very hard, in some sense, optimization problems and the solutions look positively exotic, inhuman, maybe even biological.

And it's never any slower than it is today. Celine, want to close us out?

Uh, just that it reminded me when I first saw this of the Move 37 analogy, and I think that would apply. So super exciting.

Yeah.

All right. Let's move to China. So a Chinese AI group pulls ahead of US rivals in video generation. China is winning the AI video race, not because they have better models, interestingly enough, but because they have better data. So ByteDance's SeeDance 2.0 and KaoXiao's Keling now ranked number 1 and number 2 on independent video model leaderboards, beating every American competitor. Simple reason: TikTok has generated billions of hours of video data that no US company can match. So, interesting. Let's take a quick look at a video clip here from these models. So I won't play the whole video, but the idea here, and they kind of look the same, but, you know, they, you know, according to the data, they're beating the pants off us. Alex, what do you make of this?

The rules of copyright seem to operate somewhat differently in China. And if you have access to, to more video from whatever source, however legal or illegal it is, you can train better models. I think when we were first discussing Seedance 2.0 on the pod, I made a similar comment. Utterly remarkable to me, some of the videos that seem to be popping out of Seedance 2, not because it's an seemingly an algorithmic innovation, but because for whatever reason, legal or otherwise, these Chinese frontier labs have access to more data. And there are videos floating around the internet. I think most astonishingly, there's a video floating around of a guy inserting himself into key moments in the Harry Potter Cinematic Universe. I linked to it from my newsletter, key moments, stabbing or otherwise violently intervening with of unpopular characters. I, I think probably we're going to see quite a bit of this. It, it makes the boundary between fan fiction, which historically was limited to text or maybe images, and video just completely dissolves that divide. And I, I think the, the copyright lawyers will probably have a field day if anything like this is tried in the West. But for now, it's the Chinese models that seem to have all of the both Western and Chinese video training data to do Alex, is this the place that you see China leading American models the, uh, most, uh, at this time?

That's—

I mean, it's sort of a stereotype slash cliché at this point that China was always going to have more data because they could pull more data from civilians and maybe also pull more pirated video data. And also maybe they're in a somewhat better position from an energy generation, still a weaker position algorithmically, and still in a weaker position from a chip perspective. So do they pull ahead in video generation? Right now, it seems they're pulling ahead in consumer video generation, even relative to, say, Gemini Omni, which we discussed in the last pod, for now. But maybe there will be some enormous algorithmic innovation that enables the West to again take the lead in a few months. I don't know.

You know, we, we mentioned in the last pod that Google is really the only American lab still pursuing multimodal, and we've seen all the open-source, open-weight models out of China doing multimodal. That's right. Interesting. Dave, do you want to— any comments here?

Yeah, well, I think that, you know, if you study how video generation actually works, it's using the same transformer algorithm that's caused all these other breakthroughs. So under the covers, it's the same 2017 massive breakthrough driving video generation as well. But what the Chinese have done here is, you know, you take video content and you compress it into a latent space. And then you decompress it into, into video and image creation. And within that latent space is where all the innovation happens. But other domains like chemistry, like biology, like, you know, all of physics, they also have latent spaces. And so if you see the Chinese get ahead in video generation and sustain the lead, that's a good leading indicator for every startup trying to say, I don't need to compete with Anthropic and OpenAI because I'm better at chemical reactions. I'm better at robotics. I'm better at whatever. And if you can maintain a lead at better AI within any of those latent spaces, that's a really good sign for all the startups because they can then actually maintain that lead within, you know, like Saleem within EXO, I can actually have the best company management latent space knowledge. And so, it's a really interesting, cool leading indicator. I'm kind of cheering for them to keep their lead. Which is pretty daunting versus Google because Google has all the YouTube content. China has all the TikTok and all the short-form content. But, you know, it doesn't seem like China has a massive data advantage. They're just working on this problem harder.

Yeah, I am curious how these models in terms of the speed of generation, you know, we've all talked about the notion that in the future I'm going to be generating video on the fly as it's needed. You know, sort of Netflix on demand. Alex, what's the, you know, when do you imagine we're going to see that level of video generation?

A few months ago, like world models, including GPT-4, already do that. That's what you're asking for, interactive video generation and real-time world models already do that.

Yeah, I'll tell you, we remember, I don't know if you remember, Peter, but we saw Liquid AI a year ago generate images as quickly as you could speak. Instantaneously coming out of Liquid AI. If you try to do that now today, you wait a minute or more and the experience is nowhere near as much fun as the real-time, creating as fast as you can think. But we're so short on compute and that's one of the reasons Liquid AI is doing well is because it's so much more efficient. But that holodeck experience we're trying to create is entirely possible to build the holodeck today, but nobody has the compute available, uh, to, to deliver it.

Yeah. Salim, any thoughts?

Uh, just that the— I'll echo what we talked about already, the amount of data that China has. I mean, TikTok and Douyin are huge, essentially training loops disguised as entertainment platforms, right? And so those, I think they'll stay ahead for a bit longer, but, uh, I think the models will catch up.

This episode is brought to you by Blitzy, autonomous software development with infinite code context. Blitzy uses thousands of specialized AI agents that think for hours to understand enterprise-scale code bases with millions of lines of code. Engineers start every development sprint with the Blitzy platform, bringing in their development requirements. The Blitzy platform provides a plan, then generates and pre-compiles code for each task. Blitzy delivers 80% or more of the development work autonomously while providing a guide for the final 20% of human development work required to complete the sprint. Enterprises are achieving a 5x engineering velocity increase when incorporating Blitzy as their pre-IDE development tool, pairing it with their coding copilot of choice to bring an AI-native SDLC into their org. Ready to 5x your engineering velocity? Visit blitzi.com to schedule a demo and start building with Blitzi today.

Our next story is an interesting one. I want to focus on this a little bit. Uh, uh, here's the deal. Here's the idea. The friend of Pod, Eric Schmidt, was giving a, uh, commencement address at University of Arizona. Uh, as he mentions AI, he gets booed. Um, and it's pretty brutal to watch. We'll show the video in a second. Gloria Caulfield, who's the VP at Tavistock, she's a friend, had the same treatment. The generation today that's entering the workforce is angry and scared about AI disrupting their career. Let's take a look.

No, what many of you are feeling about that, I can hear you, there is a fear We do not know. Wow.

We do not know the precise contours of what this transformation will— The rise of artificial intelligence is the next industrial revolution.

Oh, whoa, what happened?

Okay, I struck a chord.

Celine, let's go to you first on this one.

Well, I mean, look, the students are sensing that companies and institutions are adopting AI and we've not redesigned the social contract, right? The backlash is not anti-technology, it's anti-extraction. And we need a kind of a new narrative around agency, not replacement here. And there's a huge legitimacy gap between AI elites and young people today. It's like a ridiculous. Incredible to see.

Yeah. Dave Erickson.

Well, this is clearly priority one for XPRIZE now. I mean, you've got, you've got $100 billion of charitable money suddenly unleashed at OpenAI. They don't want this at all. So, I mean, but, you know, historically, no one expected this to happen this quickly. So they put very little effort. This is their wake-up call. You know, look at, look at these videos. Like, Eric Schmidt is a hero of, I mean, one of the greatest, if not the greatest business executive of all time. But when the job market is basically zero for college graduates coming out, what do you expect? You know, like, and in the sentences there weren't exactly inflammatory, right? AI is the next industrial revolution. That's not exactly a controversial— like, imagine if they'd said something controversial. What the— it'd be like a revolution in real time. But that's, that's the reality outside of the, the couple places in America where AI is happening. The rest of the country is that. And it's really important that people be aware of that. And then, you know, take that $100 billion of charitable money and get to work on this.

You know, we just launched this week the Build with Gemini XPRIZE. And again, to take a second and talk about it, we are challenging teams around the world. Google put up $3 million. It's a good start, right, to encourage teams to become entrepreneurs. Individuals become entrepreneurs, pick a problem that impacts 100,000 people and build in public in 3 months, something that generates and scales revenues. That's a real problem that generates real revenues. Uh, you know, we have after 24 hours over 2,000 who have registered. I hope we'll get to tens of thousands that register here. There's your chance to actually go and use these tools, learn to use these tools. Uh, Dan Martell, who's one of our donors, uh, you know, uh, Thank you to Dan for contributing to this XPRIZE as well. And Dick Merkin as well. You know, and Dan saying he's getting a huge amount of interest from the people. He's an incredible educator, age, you know, 13 to 25 who are going to be going after this XPRIZE. And just a shout out to the Gemini team at Google for their support on this. You know, rather than booing AI, use it to take control of your own future. Don't get a job. Up, build a job, employ other people. That's your option right now.

Can I say something more about this?

Please.

What they really should be booing is the universities that have sold them a credentialing system that is radically out of date.

Amen.

Right?

Amen.

Um, they should— because universities need to stop defending that old credentialing system and become launchpads for agency entrepreneurship. And we've talked about this ad nauseam on the pod. Um, that's who they should be booing. They've been sold a bum deal and they're in huge debt and no prospects of getting out of that.

They spend $200,000 for a degree that's worthless by the time they graduate.

Yeah.

Alex, if you're graduating, don't compound the error by getting into a job training program at an investment bank.

But just think about this for a second. If you go back to our age or even your parents' age, there's a very simple question we could ask when we all graduated, whenever, how many decades ago, which is how much of your university education did you actually use in the workplace? And the answer was nearly zero. And that was 30 years ago. Forget today. So this has been a problem that's been around for a very long time and it's not been solved and they're getting pissed off about it. And it's understandable that they're pissed off about it.

Yeah.

Well, I'll just add, I used virtually all of my education for what I do. Just saying.

Yes.

If Eric Schmidt and Gloria Caulfield were getting booed at Harvard, MIT, or Stanford or similar, for mentioning AI, then I'd be concerned. If I remember correctly, these weren't Harvard, MIT, Stanford, or similar.

No, but there's a majority of the graduating class of 2026.

What that says to me is that virtually everyone in that class is anchoring their expectations for stagnation. That's not a great place to anchor expectations at all. But also, I mean, Salim, to your point about— and Peter, your point about how they should be booing the university or booing the college, there's selection bias. You don't get to graduation if that's your attitude in general. You probably skip college altogether. So I think there's a bit of selection bias here that we focus just on those who've narrowed their possibility space while not maybe leveraging AI to its fullest, and then choose their graduation ceremony as a time to channel their angst regarding the disappearance of the lower rungs of a ladder, a professional ladder that AI is automating away, it's probably too little, too late. And they should be, I would argue, starting businesses. Inevitably, I'll get flamed for saying not everyone wants to be an entrepreneur, but not everyone wants to be a college graduate either.

There.

Speaking to the parents who are listening who've got kids in high school or in college today, or speaking to recent graduates who have not gotten a job, um, you know, this is only going to, you know, get more dramatic, and it's coming. And, you know, you need to encourage your kid, uh, to learn entrepreneurship. You know, hop into, uh, the Build with Gemini XPRIZE. Go to geminixprize.com. And learn about it. I'm gonna do that for my kids who are turning 15 next month, you know, this summer, you know, I, you know, if they should win, uh, we'll donate the money back because I'm biased. But at the end of the day, use this as an excuse to play, right? The two most important mindsets you have are purpose and curiosity mindsets. You can learn anything you want. Yeah, go ahead, go ahead.

I need to just go on a little rant here.

I love you.

So, we've been talking about purpose for more than a decade. We coined MTP back in 2012. It's not decorative. In a world of AI capability, purpose is how you orient human beings. And somewhere around 2017, I was asked to give a keynote at a conference that my team said, "You know, the logistics are really hard to work out for you to do this." I'm like, "What's the conference?" And they said, "Oh, it's a conference of 700 deans of business schools." Who knew? But they all get together. I'm like, "Hell, I absolutely want to go do that." So, I get up on stage, I'm giving the opening keynote at this event. Announcer says, "Hey, we're happy to have Salim here. He's going to tell us the latest he's seeing in EXO, et cetera, et cetera." I'm standing side stage and what you see from the crowd is completely blank looks. They have no idea who I am or what the book's about or anything. He notices this, the announcer, and he goes, "How many of you have read Exponential Organizations?" And out of 700 deans, 2 put their hands up. Now, not that everybody in the world should read the book, but if you're a car designer and the Tesla comes out, you should jolly well know what the hell the thing is. And here you have 700 deans of business schools that had no idea that there was this other paradigm out there. And I think that's just criminal negligence in a sense. Universities have been sitting on their hands for decades knowing this problem, but the immune system in academia is very strong. They're not able to get out of this. This is why we need completely new systems that ride around legacy. End of rent.

Yeah, uh, agreed. A parallel story here comes out of Stanford. Um, so a Stanford survey found 49% of 849 computer science majors would rather cheat than fail. Students say AI tools are used in nearly every class for homework, coding, and essays. Stanford brought back proctored in-person exams for the first time ever to deal with this. The honor code that Stanford was famous for is effectively dead. This is happening at the world's top CS program. Imagine what's going on in the rest of the world. Alex, what do you make of it? This is Stanford.

This is not just, you know, shows to me that there, there's a bit of an overhang for the skills that are being taught are being automated away by AI. So it's not unnatural for, uh, for these students to at least be considering using AI for all of their homework, coding essays, doubly so at schools like Stanford and like Princeton that have historically had honor codes. I've never quite understood the logic of a so-called honor code. It seems to me a recipe for laziness on the part of faculty or otherwise proctors. They should just be supervising the exams. If you want to make sure that calculators or AIs aren't being used, earn them the vast tuitions that are being paid to university and supervise. I think if anything, this points the direction— if higher ed is going to remain at all recognizable at all, and I'm not sure that it will or even deserves to, I think the direction of far more supervision, far more proctoring, maybe even some sort of wilderness camp at a university level where students are denied technologies and asked in the style of Werner Vinge and Rainbow's End and Fast Times at Fairmont High.

I love that book.

Students have to interact with—

I love that movie.

Actually the real world and interact with the real world in a way that's impossible to cheat without succeeding, like building their own actual businesses instead of writing business plans, actually solving hard problems in computer science instead of solving formulaic tests that require humans be in the room to supervise. Use AI. That's the direction this goes in.

Use AI to go 100 times bigger than you normally would.

Exactly.

Dave, you're teaching at MIT. What are your thoughts on this?

Well, I teach a class called Foundations of AI Ventures where you have to build a business plan. So it's already on exactly the mission that Alex was outlining. So, you know, if your business plan gets funded, you get an A. So it's not, it's not in this. It doesn't have this particular problem. But it seems obvious to me that you want to be teaching the students to use AI. And if you just look at their prompt stream, you can grade them. You don't have to have tests at all anyway. So, it's what Alex said that the schools are struggling to hold on to something that it goes back to when graduation speakers would show up on horseback and you'd expect the crowd to hear wisdom they couldn't hear any other way from some great human being. Why do we need graduation speakers today if we have podcasts? Well, we don't. Well, then why do we do it? Well, it's traditional. It goes back 150 years. Okay, well, you know, the exams do too. Of course it makes no sense. But, you know, it's really hard to let go of tradition after tradition after tradition. But what do you think the singularity is going to be like? You know, like, this is—

I mean, is this actually cheating or is this what students should be doing? I mean, do you expect these students not to be using AI when they get to the real world? Why are you training them for something that isn't going to exist in the future?

It's even worse. You're putting them in an impossible situation. Like, you know, forget the test, but like even in the homework, like, "No, we don't think you should use AI." Like, what does that mean? Like, you just put the kid in an impossible ethical situation. All you're doing is torture now.

Look, you're not testing durable human capability, you're testing for compliance and formatting. And that's just not the right test for this. You know, just to connect this to the previous slide, universities schools have a choice. They either become credentialing museums, right? Or they become AI-native talent accelerators. And you're going to have to make that bifurcation, and you're going to have to make it fast. Which also yields— let's go to the positive side. This is the biggest entrepreneurial opportunity in the history of education right now because the next generation of education entrepreneurs and companies, they're not going to sell you courses. They're going to sell you capability acceleration.

For the rest of your life and not just for a 4-year period of time.

Yeah, it's gonna be an ongoing partnership throughout life, through your entire life, as your education learning coaching partner.

All right, which might be here. Meta has installed software on employee computers that track mouse movements, clicks, and screen activity. The stated reason: training AI agents to understand how humans use computers. The real implications: Meta is recording everything its employees do. So AI can learn to replace them. Employees launched protests at multiple U.S. offices. An engineer's internal post about it was viewed 20,000 times. This happened the same week that Meta cut 10% of its global workforce. We're seeing this, uh, in a lot of places, and this ultimately is Elon's plan, uh, with, uh, with Megasoft. Right. Um, I mean, Mega, Mega Hard, you know, he wants to be—

Macro Hard.

Macro Hard. Thank you.

Mega Soft is a trademark. Well, that's probably—

Mega Soft.

Okay.

Mega Hard is a great name though for other things.

Macro Hard. You know, he wants to be able to come in and replace all your employees with some percentage of a GPU, um, after uploading their capabilities. So, uh, I mean, everybody figure it out. This is what Amazon is doing with its delivery workers, putting, you know, eyeglasses on them to track their movements so that future robots can do deliveries at home. We're seeing this everywhere, and it's just beginning.

You know, right now, 44% of Gen Z workers are deliberately sabotaging the AI they're supposed to train. So, so that's the, the backlash that you see there. I think this is the difference between an AI coach or an AI cop, right? Same data, but very different outcome. It depends how you frame it. And I think they need to be very careful about how they frame it. And frankly, the big issue that Meta has is there's a pretty big understandable lack of trust around what they've done. They've repeatedly said, we'll never sell user data and then sold privacy data. You can back out of these things, then you find you can't back out of these things. WhatsApp was supposed to be encrypted. Now WhatsApp is not, is taking the encryption away. It's what Cory Doctorow calls the whole inshittification stuff. It's seeing, we're seeing that now internally to the organization, not just externally in the I'll give a hot take on this one, Peter.

Please.

It seems like such a strange decision given that the frontier labs, of which Meta right now I don't consider a frontier lab. I think we have at this point 2.5 American frontier labs, none of which is Meta at the moment. The frontier labs are all purchasing enormous amounts of synthetic data for computer use assistance training across a wide variety of environments. It seems a little bit strange to me. Maybe I'm missing something. For Meta to view its own internal computer use as a valuable pre-training or more likely post-training source. It strains a little bit of credibility for, for me at least, to think that there's enough diversity of computer use just within Meta for this to be worth all of the hostility it's inevitably incurring from employees on top of all of the workforce cuts. It makes me almost wonder if this is a way of basically encouraging a subset of employees to just quit out of Meta anyway. If I were looking for better post-training data at this point for computer use, I'd be leaning heavily into synthetic data and I wouldn't be antagonizing thousands of employees with mouse tracking. That's my hot take on this.

That's fascinating. Dave, what do you make of that?

I completely agree. I was going to say a much more violent version of what Alex said. The use of this data is nonsensical as a tool internally. This is entirely about knowing what your employees are doing and using it to evaluate and sort performance. But I think people better get used to it because it's going to happen pretty much everywhere. Also, I think if you look at all the data that Google has gathered for decades now, I think a point that Eric Schmidt made years ago very quietly was that if Google ever wanted to make more money, they see every quarterly earnings report.

I had that exact conversation with him. He said we could make a ton of money just once before, before we're, you know, lawsuits start flying.

They say exactly— so that's the higher level point.

They see all the searches for all the purchases ahead, ahead of earnings reports.

Oh, and once Gmail took, took over, everyone sends their board reports out to board members. I'm on so many boards. They arrive by Gmail, you know, naked, you know, just as attachments. And Google's terms of service say right in them, yeah, well, We, human beings, won't look at all your emails, but our algorithms will. And humans could too, but they don't say they won't, but they imply they won't. But the algorithms look at them. So yeah, but I think the high-level point there is that the data gathering is happening for sure. And you're crazy to misuse the data because of the backlash. And this to me is what Alex said. You just created a negative news report that you didn't need. Why, why do that? We know you're gathering the data, but don't, don't abuse that gathering in a way that creates some massive uproar like this. Cause that's what's been going on in China for, for decades, but it's also Google has all the data. Apple has all the data. They don't really look at it. Um, so the data gathering is happening and Peter, you've made that point over and over again. So don't play— getting in an uproar and fighting it is crazy though. You're just labeling yourself as a difficult-to-hire person. It's not going to stop because of your, like, having a rant about it. So I'm not saying, you know, just adopt it either, but I'm saying don't, don't just rant and then go home and feel like you did something. You didn't achieve anything.

Tying a story here back to the conversation about job loss, Mark Cuban proposed federal token tax. This is his quote. Quote, we should tax token at a provider level, less than 50 cents per million, as it will push big AI players to optimize tokenization. We'll also reduce energy usage and generate $10 billion a year, maybe growing 30x or 100x, while also creating a source for paying down the federal debt. You know, uh, we've talked about the idea that we're going to potentially tax, uh, you know, AIs that replace employees or robots replace employees. But these numbers here that Mark puts forward are de minimis, right? This is for 300 million Americans. It's like $33 per year. I mean, interesting idea, but I don't think this moves the needle anywhere.

And certainly, I mean, so there are so many perverse incentives when one introduces novel types of taxes or floats taxes like this. The first obvious novel incentive is get rid of tokens entirely. There are many token-free or tokenization-free approaches to autoregressive and non-autoregressive machine learning at this point. We could swap to diffusion models entirely if we wanted. There are ways to do diffusion models with no tokens, or we could skip the tokenizer step entirely and stick with transformers. All this, like naively, all this would do is just push for the abolition of tokens altogether. I think then the response is, okay, fine, let's tax the flop. Let's tax floating point operations or something there, more perverse incentives. I think it's very difficult to come up with an input resource that would be best taxed short of actual dollars or some function of dollars that would actually enable a pure play targeting of the AI labs without creating severe perverse incentives like this?

I mean, Alex, you remember that Sam recently talked about, well, we should have the general public own a part of national compute, right? This is the equivalent of the Permanent Fund in Alaska or what occurs in Saudi, the Emirates. You know, as all of the capital starts flowing into the agentic ecosystem, into these frontier labs, and they start generating huge amounts of wealth, you know, how do— how does part of that wealth get distributed to the average American?

I think we're going to start testing that. I think we saw it in— so I'm not sure if we have a slide for this, but he's testing that right now in the form just in the past 24 hours of offering $2 million of OpenAI tokens in return for a safe to all YC companies, like 2,000 companies. I view that as a preview of some universal basic compute-type offering to everyone. Not quite obvious what the equity equivalent is, but something like that.

Yeah. Dave?

It's amazing how quickly we've decided that LLM AI and the word token are a fixed thing that can be taxed like bananas or shipping containers. But anyone who actually works in the industry knows that You can, you can actually tokenize with byte pair encoding, which is what you usually do, but you can use 1-byte encoding. You can use 4-byte encoding. You can use no encoding at all. Like suddenly what you tax disappears from the world like a minute later. Uh, so yeah, it's, it's just funny that people feel like I've got an idea. Let's tax per token. Like it reminds me of a conversation in Riyadh where one of the royal family, uh, that we were meeting with was, was, um, saying, what do you think a good cost per token is? Cause we're big investors in a couple of companies. And their cost per token is like a buck a million. I'm like, with what context window and what— that's a nonsensical question. And they, they're like, well, we'll just answer it. I was like, so I gave an answer and they're like, great, because we're less than that. We cost less.

Okay.

Yeah, that's going to be really tricky and amorphous and fast moving.

Look, it's interesting at one level because it treats AI usage as an economic activity, right? It goes to Alex's comment from another pod where we talked about what's the economic return per token. And, and that's really interesting because it's not, it's not just a software feature, right? Now, historically, governments always taxed what society's dependent on, whether it was land or labor or trade or income. If AI becomes a new engine of productivity, then tokens or some flavor— compute, energy— become obvious tax targets. The challenge is that if you try and tax it, then there's nothing more liquid that compute usage today, it'll go to people that don't tax it. And you risk doing that. The other, I think, big danger is that you may put an inadvertent fence around Meta and OpenAI and the ones that can navigate this easily and stop, slow down innovation because startups will have a harder time around this. So there's a whole bunch of different decisions here, but I think the flaw in trying to tax tokens itself, the intent is right, but the mechanics may not be Yeah, but I think you've already like laid out the portrait for, for maybe Neal Stephenson's next cyberpunk novel.

It's got to be about like compute tax havens on Peter Thiel-funded like data centers, seasteads on the ocean to avoid all the token taxes.

We should just list all of the science fiction novels with the number. And in each episode, we just list out here the 45 science fiction plots we talked about in this episode.

Bingo.

All right, here's a fun conversation. Uh, I'm a huge fan of Ben Lamb, George Church, and Colossal, and they just had an announcement that I wanted to share with everybody. Colossal Biosciences hatched chicks with an artificial egg. Uh, let's take a listen to this video. Begins with the chicken egg. We didn't want to just reimagine the egg, we wanted to completely re-engineer Now showing the world that we can actually grow this whole bird now in an incubator outside of an eggshell. Complete game changer. Unlocks massive potential.

I am looking at the world changing and I'm holding it in my hands. I think that this is really a big step for science as a whole, honestly.

Hold on to your chickens. This is the Colossal Artificial Egg. As we developed the design, we began with a natural egg.

So the Colossal Artificial Egg is a world's first in terms of the design level and sophistication.

The first major structure is a rigid outer shell. The shell provides the protection, rigidity.

Our permeable membrane allows oxygen to diffuse into the system through the membrane at ambient temperatures.

It's got a really big window on the top that you can actually really look at and see and understand exactly what is happening to that embryo.

Real-time visibility into every stage of embryo embryonic development.

This is a real breakthrough.

So this is ex utero gestation. Uh, this is one of 3 artificial womb programs that Colossus has going on. Remember, Colossal— Colossal has going on— Colossal is the company that's bringing back extinct species. They're famously— the first one is the woolly mammoth. They brought back the dire wolf. They're bringing back the dodo bird. They have 15 different species in process, but it turns out like the MOA, um, uh, is, is got a huge, uh, egg requirement, uh, and just doing it, uh, is going to be challenging. Uh, super curious, and just for the numbers, there's 1.9 trillion eggs per year, uh, that's generated by about 33 billion chickens. So interesting. Interesting. Thoughts on this one, gentlemen?

I'll just note, I can't wait for the dodo, which I think is one of the targets for this. They've tried this now with a couple dozen birds. I think it's a far cry from what we saw in Jurassic Park, where the emphasis in, in that Mr. DNA reel was more on the molecular biology of, of resurrecting extinct species and not organism level or egg level focus. I think it's, it's important. And as I understand it, there have been some major challenges that prevent the naive Jurassic Park type approach from working in the case of, uh, of eggs for these extinct birds. Apparently it's the case that late in gestation they require large amounts of oxygen that are difficult to supply through artificial eggs. And so I at least have historically have been difficult to supply. So creating, uh, effectively, uh, an oxygen permeable artificial egg that's supportive of late gestation metabolism seems like an important step. And the, the, the giddy sci-fi fan in me wants to know how well will some of these techniques generalize to humans? Can we perhaps appropriately generalize these to ex utero gestation in humans?

They're working on mammals, um, and obviously, uh, you know, avian and also, uh, uh, you know, bringing back, uh, I think of the 15 species in the pipeline, two-thirds are mammals right now. So yeah, the dodo bird, which is native to Mauritius, or was native to Mauritius, uh, on their money, on their stamps, on their flag, is extinct. And, uh, the country's very excited to bring it back, drive tourism revenue you.

Well, at least 2 of the 15— the dodo and what was the deer, the, the red or blue deer?

Some— yeah, the bluebuck, I think, or something like that.

A bluebuck.

Yeah, so those 2 were extinct because they were delicious and easy to hunt. So Salim is opening dodo fillet restaurants, uh, getting ready, and maybe a burger, a bluebuck burger.

Yeah. Anyway, I just wanted to, you know, this is the This is the science fiction future materializing before us. We're not going to have, uh, uh, you know, dinosaurs per se, though what, you know, Colossal— what Colossal is doing is not actually bringing back an exact species. It's bringing back a singular simulacrum. It's being able to say, we're going to make these 300 edits in the genome that will make this species look like this. We'll have longer tusks, so we'll have have woolly hair, it'll be more cold tolerant, it will have a longer snout. And so you can sort of design your species. I had Ben Lamb on stage in, uh, in Miami at FII and said, you know, can you, uh, can you create a dragon? And the answer is yes. Probably not fire-breathing, but we can add the wings and we can make it look like a dragon.

Um, yeah, I think, well, I mean, dinosaurs, Peter, I'm pretty confident we'll get dinosaurs.

Will get some version of a dinosaur.

You know, I don't know if you remember, but my visioneering, uh, proposal a few years ago was that genotype to phenotype mapping problem where you start with a picture of the final outcome and then you try and create the DNA that matches the picture.

Exactly, that's what they do.

Yeah, yeah, they're building—

you could, you know, you could blind test and see if, you know, one theory is you created a completely different animal, it just looks like the original. The other is no, the only way you can create the original is by getting the DNA within what could have been, you know, the reproducing line of that original species. No other DNA is going to get you to that exact look, feel, behavior, shape, whatever. And you could backtest that and get the answer to that.

We're going to have, we're going to have Ben Lamb with us on stage at the Moonshot Gathering in September. So you guys can drill in on this. He'll be one of the Moonshot entrepreneurs we're going to be bringing with us.

Be ironic if they could resurrect stochastic parrots with stochastic parrots.

Uh, no one's going to get that.

You don't think so?

Um, someone in the audience will appreciate it.

There's a thread that's been unraveling for a while and it continues to do so, which is that biology is becoming programmable. Yeah, that's just a huge, huge, huge thing.

Well, he's working on not only biology for animals but biology for plants, being able to design a plant that is drought resistant, disease resistant, can grow twice as fast or twice as big. I mean, this is the intersection of AI and synthetic biology. Um, yeah, very exciting times, everybody. Welcome to the health section of Moonshots, brought to you by Fountain Life. You know, we talk about AI on this Moonshot podcast all the time. One of the most important things AI is going to be able to do for you, besides educating your kids and helping you with your taxes, is making sure that you're living a healthy lifestyle, that you get a chance to get to 100 I'm here today with Dr. Dawn Musalem, the Chief Medical Officer of Fountain Life and a part of my medical team. Dawn, a pleasure.

Great to be here.

You know, the thing that people are concerned about most about living to 100 or 120 is their cognitive abilities, making sure they don't have dementia. And the numbers about dementia are problematic. Can you share what you've learned?

Such an important point, and you're right. Our members, the number one thing people are most concerned about is losing their brain health. Forgetting the name of their child, forgetting the face of their loved one. We know that when it comes to dementia, the conservative estimates are that 45% are entirely preventable. What was amazing is with the advanced testing we're doing at Fountain Life, one quarter of our members had advanced brain age.

Wow.

But what was really awesome is, again, back to that prevention, when we partnered it with Healthy Living— this gives me chills— individu— eating healthier, moving our bodies, sleep. Optimizing sleep is so important. You know what we saw? We saw that we improved that brain age by 26%. That is a big, big number to show that the majority of those individuals were able actually to improve the brain age.

And one of the things I love about Fountain is we're searching the world for the best therapeutics, the best approaches, and making sure we bring it to our members. So if having healthy brain function till 100, 120 is important to you, check out Fountain Life. Go to fountainlife.com/peter. Make sure you become the CEO of your own health. All right, now back to the episode. All right, 3 stories on data centers and energy, uh, worth noting. Uh, let's take a look. So a Gallup poll found that 70% of Americans oppose data center construction in their community. Nearly half strongly oppose them. Some residents said that they'd rather live near a nuclear plant than a data center. Main concerns: rising electricity costs, water usage, environmental damage. This is a not-in-my-backyard problem that needs to be addressed. You know, we've just either stopped or delayed 7 gigawatts of data centers, nearly half of the data centers being proposed. It's a real challenge. And one of the questions is, you know, who's, who's funding the protests? Who is, uh, you know, giving these individuals the information? Um, I don't know. So comments on this, gents?

I mean, this is— this ex-prize should pick this up immediately and say, okay, the information is nonsensical, and this is proof positive that people will freak out about anything that they don't understand that's a change that's coming to their neighborhood. And so the solution is some combination of PR and education and making it a smooth thing And, and it's just such a solvable problem if you put some budget and energy and thinking behind it. But, you know, these act like the idea that you would rather live next to a nuclear power plant instead of a data center, which you would never even know it was there. It's not taking your water. I promise you, it's not taking your water. And the electrical effect, they already said, like, you've got to find your own power. You can't disrupt the local power. Like, there's nothing in those objections. But if you've been to a town hall meeting, you know, there's a group of people people, they, they will object to anything that's brought up any given day. That's what they do.

Let's get the data here for the next story and then, and then talk about it. So, uh, Nevada Energy is redirecting 75% of Lake Tahoe's electricity supply to data centers by 2027. Um, you know, these are the numbers that are projected, uh, and of course we've talked about this. We're going to have Michael Kratzios on the show. We'll be talking about policy in this area. Uh, you know, we have solutions here. We have the data centers being able to stand up their own supplies. I mean, I think that the hyperscalers need to be saying, listen, if we build a data center in your backyard, you get free electricity, right? That would be the solution to turn the tide here.

Well, this isn't what— what's actually going on here is different from what the story implies. What's going on here is California in November is going to vote on whether to take away 5% of the money of 200 billionaires in California. Uh, just not their income, but we're just— whatever money you've made in your life, we're going to take 5% of it as a one-time tax. People are leaving to go to Incline Village, Nevada, by the ton. Nevada doesn't have much of an economy, but they have a beautiful shore of Lake Tahoe where all the Californians are snatching up houses at an incredible clip to try and get out of the state. So Nevada is now saying, we like data centers because we like the inflow of money and success this is going to bring to our otherwise desert state. I mean, this makes Vegas look like a rounding error compared to what we could achieve with this. So you've got this cultural divide between the California protect everything view and Nevada saying, whoa, what an opportunity view. So no one's going to drain Lake Tahoe. That's not really what's going on. This is state versus state, completely different perspective on which way to go with compute and AI and data centers.

Alex, what's your take here?

I've commented in the past, over the past 2 to 3 episodes about how the tokens want to flow to the highest dollar per token productivity ratio. I think we'll see in the early lead-up to the Dyson swarm, the, the kilowatts flowing to the highest dollar-per-kilowatt-value applications as well. And for those in the audience, there will inevitably be someone who takes issue with the story and says, no, you're misconstruing, this has been in the works since the late 2000s, since 2009, it's not actually an AI data center story. It is actually an AI data center story. Story that there— it's a bit inside baseball, but based on public reporting, the transition away of some of the funding from NV Energy to the local electric utility for Lake Tahoe has been planned since 2009 to be switched off or switched away from Lake Tahoe, but has been perpetually extended. And this time around, it looks like it's not going to be extended because there are data centers that could be far more productive consumers of that electricity supply than the residents of Lake Tahoe. So I view this ultimately as a kilowatts— until we build the Dyson Swarm, the kilowatts— or radically expand our energy supply. And we're probably, I think, going to do both at the same time. The kilowatts want to flow naturally to their highest productivity outcomes.

Well, speaking about expanding energy, here's a story that I want to show these charts again. It's part of our abundance story that we've talked about in the past, that Texas Texas surpasses California in utility-scale solar. And, and Elon, you know, friend of the pod, has made this comment a number of times: we can get all the energy we need from solar. Uh, here are the numbers on utility scale. We see in this chart here Texas, uh, doubling the amount of solar in Texas, uh, surpassing California over the last 5 years. Um, we see Texas, you know, again just about doubling the amount of storage, and Texas, you know, going— what does it look like— almost 8x the amount of wind. Uh, you can build renewables in your state. I mean, every part of the U.S. has some version of renewables it can put in place. Uh, and I just don't see why we're not doing more of this. All right, you know, China has done laps around us here, but good on for Texas.

It's quite the indictment of the California state government, doubly so given that Texas has all of these amazing oil and gas resources on top of it all.

Good point.

That's a permitting story.

Yeah.

Yeah. Well, and one of the things that makes America strong is that we have 50 states that, that actually can compete. And the more we, more we allow them to compete, the better off we are in the long run. You know, California right now has incredible tailwinds. Like, I can't even tell But historically, the legislature will say, "Ooh, more tailwinds, great, more taxes," and just keep absorbing a bigger and bigger fraction of the growth. But I don't see anything slowing down California's AI success. It's just off the charts. But Texas is doing a great job of taking control of the data center side of it, and they're going to benefit tremendously from that.

And the data centers can live wherever the energy flows. I think that's the point, Alex, you brilliantly made. That's right. We're going to see geothermal. There's lots of geothermal hotspots around the country. We can go mine, and we can build solar. I mean, God, every desert out there deserves solar farms.

Right now, Texas is the closest thing the United States has to a special economic zone. Maybe this new agreement in the Philippines that the White House has made, maybe that will become the American Shenzhen. But short of that, Texas is our special economic zone. So I'd say we should make the maximum amount that we can from this federalist system and use it.

Nice. I mean, energy is becoming the balance sheet of, of AI, right? It's, it is very clear.

It's a key input. It's not the only input, but it is an input.

Yeah, it's the limiting input for a lot of, uh, of the growth right now. All right, we're going to turn to our final story here, uh, is the organizational singularity with Salim. So Salim, you and I recorded, I think, an extraordinary hour-long conversation about the singularity party, uh, and I encourage people to go and watch it. But those of you who haven't or have or won't, uh, I want to give Salim a chance on this pod to present his thesis and then have the mates, uh, kick it back and forth. So Salim, over to you, pal.

Yeah, so, you know, we've had 80 years of thinking about the organization. Coase said the transaction costs and coordination costs are cheaper inside rather than outside. Then you had people thinking about, okay, how well do human, human teams make decisions. Then we had Clay Christensen come up with this innovator's dilemma. Then we had lean startup thinking. Then we had platform business models. We had EXO 1.0 that took the Kosean firm and stretched it outside the boundary using community and crowd and AI. But in the face of AI, all of that breaks, right? The best— the cost of doing any transaction inside a company is more expensive than doing it outside a company, and the metabolism of almost every company in the world is slower than the outside world. And so, essentially, Koso's Law essentially breaks because of the externalities have broken. The famous tweet that I keep remembering is, "It's easier to build a product feature than having the meeting about building the product feature," right? Which just nailed it for me in one thing. And so, in this world, and yet 80%, if you can go to the next slide, Peter, we're seeing this totally change. We need a totally new architecture for what an organization looks like. And we've come up with a kind of three-fourfold shape here. At the core of it, you have your MTP, which actually becomes a protocol for how you think about this, and then an intelligence engine, and then an organizational form around that intelligence engine. I think the next slide may have like a rocket ship picture of this. Well, let me go to this. The heart of it is an intelligence stack. And as we— after we built this, which is like a sensing layer and orientation layer, we realized that this is essentially in OODA loop, right? And we've seen this in the military world where you can sense, organize, react, and with a feedback loop, and you need very strong signals around this. Because in this new world, even the EXO model doesn't work. You need a totally different architecture. So if you put an intelligence stack at the core, and instead of organizing around hierarchy, you organize around intelligence, you now have a proper wrapper for what 21st century organizations should look like. The boundary of the firm changes. It used to be that we had legal entity called the organization that served primarily for execution and coordination. But, but if you can now do execution and coordination automatically, or rather magically with AI, then the firm becomes a purpose container, a fiduciary container, a legal container, like a glorified SPV, essentially, a liability container, but not where execution gets done. Because AI agents are going to make— be making API calls all over the place. And so you need this stack which learns constantly with a very strong governance loop around it. And I think we have it laid out in the next slide a little bit, what that looks like. You need for every agent very trusted eval suites, you need searchable logs, you need a rollback capability, you need a human review queue. This is where, as we're watching AI agents propagate, it turns out you need a huge amount of human oversight because they're, they're gonna go— Martin Wersowski talked about this, they're like junior employees, they go rogue pretty easily, you have to watch them very carefully. And so you need a very strong governance architecture around all of these. So you have a totally new architecture around this. And the problem today is that 80%+ of AI projects are failing. And the reason that they're failing is we're taking existing AI and trying to cram it into human-centric workflow loops. And all you're trying to do is automate the human-to-human bottleneck. Of course, it's gonna fail. You need a totally different model. And so we've been at, for a decade+, we've been going down the old trope of you have to do disruptive innovation at the edge of the organization. You can't do it in the core organization. If you can go to the next slide, Peter. You have to do it in a very— oh, okay. So you have to do it in a very different way. The only way to do it is to rewrite your organization, create an AI-native digital twin at the edge, and then move workflows over one by one, red team it for a while so that you're not threatening the mothership. And once you have recursive self-improvement at the workflow level, you can start to slowly deprecate the old human beings become oversight, dashboard monitoring, exception handling, problem solving, et cetera. And we're gonna need a huge amount of oversight. So I'll give a quick analogy here. Imagine you had a, you're running a trucking company and a competitor announces refrigerator trucks as a new line of business. So first you have sensing agents that tech figure this out and go, hey, they bring back information that, hey, competitor's launched refrigerator trucks. Now you have a layer of strategy agents 'cause strategy cannot be a static process. It's gotta be a living protocol of saying, okay, how big a deal is this? How big is the market? Should we think about this or not? And evaluate whether this is a strategic option. The next layer is kind of an analytical layer that says, okay, if we were to do this, there's kind of 2, 3 ways we could buy a refrigerating trucking company. We could launch our own trucks. We could do this or that. And comes to a decision layer. Do you buy a startup? Do you launch your own trucks? Do you lease some trucks as a pilot? Maybe you say you want to launch, then a set of execution agents goes and does that. What are the human beings doing? They're hitting approval at every point and slowly letting the AI run most of this. And instead of a typical C-suite taking weeks or months to make that choice, you're doing it in days, right? When you, when you have this kind of AI-native architecture, our current assessment is that you should have an organization that's between 100 times or more performant than the legacy. And so that gives you a huge incentive. And we can see, what was that company that got 73 times ARR in a year by running AI native, right? We can see this happening. We've seen this transition fully in call centers that went from human to chatbot assisted and then AI native, or marketing content, which was agency driven, then AI assisted. Now most AI content is AI-driven. So we've developed an entire methodology we call Rewrite to help companies create this digital twin at the edge. What we're going to do is take batches of CEOs and run them through this process. So if you're interested, watch the— go watch the longer video where Peter and I go through this in a bit of detail. And then you— we give you a process and an AI. So we're releasing this whole book as AI because every 2, 3 days, the world changes. We have to update it, like, to version 15.643 will be released. And, and we're going to launch it as an AI, take batches of CEOs through this. We're also launching, by the way, an Organizational Singularity Venture Fund that goes along with this because companies may want funding to either build a company in this modality or for doing it this— in this new way. So if you're interested, I'm happy to discuss it further with the pod base here.

Where do they go to get more information?

Go to organizationalsingularity.com. It'll point you back to a page, uh, companies can apply to go through this process. I think we've got, uh, 4 of the initial 15 slots filled. Um, the largest education group in Brazil is going to run a bunch of universities through this process. Interestingly, that's a a big kind of category because they need to adapt and change. But the core thesis is the way we've been building companies for 100 years now completely changes to be very centric. In the book, we actually have written a whole chapter on how, as you guys have done your whole logic for modeling the future, right? Solve everything. But what's the organizational design that allows you to take domain after domain and bring about domain collapse in that? So we've got a whole chapter saying how does this work for that? There's a chapter architecture we're putting in on governments and nonprofits because there's an obvious— most government processes are very prescriptive and lend themselves well to this type of AI-native architecture. And so we've built, written all this out. We're gonna keep evolving it as things change and as models change and as agent architectures change. We've compared this to all the writing. When we wrote the first EXO book, Peter, I think we were probably 7, 8 years ahead of market. The second one, it was maybe 3 years ahead. I think this one, we're maybe a year ahead. So it's much more timely than the other ones, which were way further in the future. I think the simplest way to talk about this is, is you have, you don't have a choice. And here's the question that every CEO and C-suite and board member needs to be asking. Can 2 guys with Open Claw replicate a major line of business, a high margin line of business that you have in in 60 to 90 days. If that's the case, you have an existential threat right now. And so you have to do something like this to move to this model.

All right.

Let's take a couple of questions from the mates and yeah, Dave or Alex.

I'll bite on this one. So Salim, I'd love to push on this a bit and from the perspective of falsifiability. So you mentioned KOS earlier. I've written a fair amount at this point in the context in the context of some of my own investments regarding coasts are arguably the intercontinental railroad, for example, following Kosean economics pushed the firm size to be larger. Now you needed huge continent-spanning companies to encompass intercontinental railroads. I've argued that AI pushes the firm size smaller because maybe you can, as I've argued in past, you could have one-person conglomerates. Can you make a falsifiable prediction given all of this about the size of the firm?

It can go either way. We're writing— I've mentioned in other pods, we're writing a whole economics technical paper that, well, how would you measure the size of firms? Because the size is not that relevant. It's the amount of economic throughput you can put through the organization, right? And that will depend on domain. It'll depend on what kind of moat you have. So for example, regulated companies will have a— will be able to defend themselves a more. If you have proprietary data, you'll be able to defend a lot more. The best defense will be if you have an inner intelligence loop, per your thinking, Alex, right? Once you have that, that's very defensible, and you now have a flywheel that keeps building. So we don't know what the false survival model would be. Certainly for some entities, like very customizable work, if you're building a, a sports kit car business, and that's very manual and labor-intensive, etc., uh, that won't fit model very much, at least early on, but we'll definitely get to that point over time. But we think that this will apply overall as you replace human-centric hierarchical coordination with an AI-native loop, and then that will just keep flywheeling away.

Let me try maybe an adjacent question, if I may. So you're laying out, as I understand it, a theory of the future of the firm or future of organization it sounds like it has both descriptive and normative, in other words, what should happen versus what will happen or vice versa, components to it. How should an organization that is intrigued by this, how should it judge whether this is at least an accurate description of the future of the firm? How do we quantify this to know whether this is actually— like, what are your benchmarks? How do we know whether this is even accurate?

So we have early signals from the market, right? You can see the rise of some of the AI-native companies. Klarna has done an amazing job of turning their customer service environment into a profitable center, etc. Cognizant or Cognient or whatever that was. You've got companies like Pulsia that are creating an entire scaffolding and launching companies without blinking. It's not so much the number of people as how much economic throughput you can put into it. But I think the other way to answer this question is when we launched the EXO model, we measured all the Fortune 100 against it. And then 7 years later, we found the more— the top 10 of the Fortune 100 that followed the EXO characteristics the most delivered 40 times the shareholder returns of the ones that delivered it the least, that used the model the least, right? Because why? Because as the external world becomes more volatile, your ability to adapt will drive market value. We totally should have set up an index index fund. It would've been the highest performing index fund ever. In the same way, given how accurate we were that time, we're basically going on that reputation and saying, look, we were way ahead of our time, but we nailed the architecture, we nailed MTP, purpose-driven organizations. Now we're seeing companies hire based on the alignment of a personal MTP with the organizational MPP, otherwise why bother, et cetera. As we drive this forward, I think the broad thesis of intelligence being at the heart of it is a very defensible one. We may Maybe have to adapt some of the attributes around it, but you need a trust architecture. You need very proper scaffolding around the agentic harness to track what they're doing, audit trails, et cetera. And then we think that general direction is totally defensible. We'll have more and more examples over time. There's an insurance company that we saw that is using 2,500 agents and that's doing the work of like 500 people. Okay? And so there, we'll start to document these and do tear-downs. So we're launching a YouTube channel called the Shift was going to go through and go, go through this in detail and interview people that are doing this and bring out signal from noise as quickly as possible on this overall architecture.

And Salim and I do an hour in which I think it was one of the best conversations I've had with Salim on this topic, walking you through what it means, how to implement it. And honestly, it's a, it's an hour-long conversation I'm going to be sending to the CEO of every company I'm involved with. So Salim, uh, thank you so much. Uh, I want to thank everybody for listening. By the way, we're at 499,000 subscribers. Are you going to be the 500,000th subscriber? Hit subscribe now, turn on notifications. We're now beginning to put out 2 podcasts a week on a pretty regular basis, uh, living with my moonshot mates here, because I feel completely out of the loop if I'm not having these conversations with guys.

Me too.

It is awesome.

It is awesome.

And I'll be in the innermost loop, right?

Yes, it is. All right, I'm going to play this outro from Simon Garrity called Who We Are. Let's listen up. Some good energy here. Rather than inevitably leading to dystopia, AI could catalyze spiritual growth and global harmony as Machines replicate aspects of human cognition. We are compelled to reconsider what it means to be aware, to feel, and to exist. AI exposes biases, flaws, and inequalities in human institutions, revealing patterns that are often hidden. This reflection can be uncomfortable but creates opportunities for collective self-correction.

Very nice.

Machines also challenge the human ego when algorithms All right, gentlemen, uh, always a pleasure to be, uh, spending time with you. Um, anyway, I could not be more excited. Uh, it really is the best time ever to build. Everybody go and build, be an entrepreneur, jump in with your LLM, tell them who you are, what you love doing, your purpose in life if you know it, and ask, you know, what are some business ideas and how I go for— you can program in English today, you know. Join us, uh, at the geminixprize.com website and register. All right, uh, Salim, good luck in Brazil. Dave, enjoy the Bay Area. Uh, Alex, enjoy wherever in the virtual world you might happen to be.

That's just my meat body, Peter.

Yes. See you guys soon.

All right, folks, see you See you soon.

Thanks, Peter.

If you made it to the end of this episode, which you obviously did, I consider you a Moonshot Mate. Every week, my Moonshot Mates and I spend a lot of energy and time to really deliver you the news that matters. If you're a subscriber, thank you. If you're not a subscriber yet, please consider subscribing so you get the news as it comes out. I also want to invite you to join me on my weekly newsletter called MetaTrends. I have a research team. You may not know this, but we spend the entire entire week looking at the meta trends that are impacting your family, your company, your industry, your nation. And I put this into a 2-minute read every week. If you'd like to get access to the Meta Trends newsletter every week, go to diamandis.com/metatrends. That's diamandis.com/metatrends. Thank you again for joining us today. It's a blast for us to put this together every week.

Study and play come together on a Windows 11 PC, and for a limited time, college students get the best of both worlds. Get the Unreal College Deal, everything you need to study and play with select Windows 11 PCs. Eligible students get a year of Microsoft 365 Premium and a year of Xbox Game Pass Ultimate with the custom color Xbox wireless controller. Learn more at windows.com/studentoffer. While supplies last. Ends June 30th. Terms at aka.ms/collegepc.