Raoul Pal: The Economic Singularity Is Coming by 2030 — Here's How to Position Now

AI is easily the greatest discovery humanity will ever make. >> This is not an easy game to win or to start, but this is not going to be stopped. >> AI is sucking up all of the liquidity and it's causing a problem in crypto. And now, oh Jesus, we're 2008 again. >> This has cost more opportunity cost than anything I've ever seen. >> Meanwhile, the dollar keeps going up and they all say, "Yeah, but it's going to crash." >> Every empire that's ever had a reserve currency has lost it. >> What useful information is that? How useful How useful is it? How does that help anybody apart from create a long doomy story about the world so the US has lost its empire, the dollar is over, it's all going to collapse. How does that help anybody? Who do you think is going to be running economies in 10 years time? It's obviously going to be AI. >> What I see happening in crypto shows that AI, a good thing, can pull liquidity out of the system so profoundly that somebody that was in crypto on debt now finds themselves in trouble. If we get a bearish sentiment that ends up hitting let's just round it to the eurodollar. And people just suddenly start going, "Nah, I don't want to lend money anymore." To whatever. And now, oh Jesus, we like we're 2008 again. >> But Tom, let's just call it a business cycle. Happens all the time. It's exactly That's what it's supposed to do. >> So you know something like that's probably going to happen, but just don't be a dumbass. >> It's very clear to me by 2030 we have no clue how the world's going to work really. We just we don't. >> That's the economic singularity. >> Yeah, we just don't at that point. What does it mean when you put an AGI brain into a robot? You know, what does it mean when GDP growth can go at 20%? Who does it accrue to, as you say? To the invisible economy of agents we can never see, we can never touch, we can't talk to. That's a wild world, right? So, I I've been saying, you know, all along, I've been very consistent, is this period is pretty easy. We've got them the exponential age of technology kicking in furiously, which I started talking about about 5, 6 years ago. And we're like, we've never seen anything like this. And to your point, the Central Bank has to keep debating currency, so our investments go up, regardless. And we're in a race with China that we cannot lose, simply cannot lose. It's humanity level stakes. And China will say the same thing, they cannot lose the US cuz no, we can't have a world system run by one AI. Can't happen. So, therefore, the chances and we got the CapEx spend, the biggest in history, planned out by companies with huge amounts of money. So, the chances are this elongates for a bit. Now, the problem with it elongating is you end up with more ex- excesses. Because not everybody's great, right? You end up with people misvaluing all sorts of shares, all sorts of things, things can go bust, whatever. I just don't think it's now. I just don't think it's now, you know? Um and then I don't know what the world looks like in 2030. What is a company? What is a corporation? It's to give a a group of people together the rights of a human, which is what's corporate, so it has the legal rights and status, and it exists to make a profit. I don't know, it's a weird concept actually, the bunch of humans working for this entity that is no thing. Like, what is Royal Dutch Shell, one of the biggest oil companies in the world? Nobody knows anymore what it is. It's just a thing that just does its thing. No reason that can't be an automated thing doing its thing. So, what is the value of half of these companies? We don't know. What could the stock market do in 2032? Well, you know, SpaceX could be worth $10 trillion and and 80% of the US companies could fail. And the economy could still grow. It's like it's a really weird world, Tom, and and I don't know what it means. And so, I just think I'm not going to worry about that yet. I'll worry about trying to protect yourself by making as much money as possible in the interim. Um and make intelligent decisions by using it as an opportunity to not have that. To be able to observe rather than freak out what it's going to do to me. You can observe and say, I wonder how this is going to play out. It's a much easier mindset to go through if you've kind of planned for it. And this is exactly what I've been talking about. You know, when I started, I said, you have 6 years to unfuck your future, then 5 years to unfuck your future. We've got about 4 years to get this right. The gift of the opportunity was there. I mean, technology, we're crypto before it. Now we've got technology, crypto probably have another one. It's all there. Just invest in your own demise. Don't fear the demise. >> Wow. >> you if you always think that tomorrow's going to be a rainy day, you never go out anywhere. Or today, you know, you just And it stops you investing. And that's what I've I learned the hard way that I always looked for the doom. And what I did was miss the upside. And not over short periods of time, but I I still see it with friends of mine. The kind of gold bugs who've been doing this doom-mongering for 30 years. And they've not made money because they haven't been able to invest. They've kept their capital You know, they they've they've staved off devaluation and and um debasement by owning gold, let's say. But they've missed the entire output of humanity, which is this increasing in intelligence. They hated it because it was always going to go bust. But that's mad. When human ingenuity and the rise of intelligence and the increase in the value of Nasdaq companies or technology companies overall is relentless. The technology companies of the past you know, get replaced by new technology companies. So, yeah, could you be in the wrong company? You just buy the Nasdaq. It's the same thing with the crypto. Always buy Bitcoin, buy Nasdaq and save up some cash and if there's a big sell-off when it's down 50% you say thank you very much and buy some more. >> How does somebody avoid the um reality that when the dot-com bust happened um it took 15 to 20 years for a lot of those companies to rebound. >> So, you're missing what I'm saying. Let's say price of asset peak is 100. Let's say you got in at the last part of it. So, you got in at 70 or 80, let's say, right? So, you didn't get the absolute top and it falls 90%. So, it's down to 10. >> Yep. >> And you double your position. Your average is now 50. Where did it How long did it take to clear that? Not very long at all. If you chose a bit of stock picking, had you chosen Amazon into that fall, which had been down 96%? I mean, Christ. The amount of money you would made, you would more than compensated for that sell-off that you had. But people just think so short term. Because what they're not doing is projecting their future self. Here, you're projecting your current self onto your future. So, if you're saying, "Hey, there's a secular trend that goes probably into infinity and there'll be some big ups and downs on route and I'll take advantage of those by if it gets wildly overpriced, I might sell some and if it gets back down to super cheap levels, I'll just buy more. But most people actually fear going there because they're worried there's a pothole in the road. It's like I say I don't get it. Of course I get it because the human emotion side is the hardest part and I've been there. I've been paralyzed by that stuff in the past. I I was paralyzed from 2008 to really buy equities after 2008. I just didn't buy equities. It took me until 2020 to buy equities again. Yeah. I mean I made money in all sorts of bond markets and currency markets and stuff, but equities, no way. And in fact, I would I saw 2000 coming because I saw this demographic bust coming as well. So I hadn't been long equities myself personally since 1997. >> So wait, this seems like you're um now I'm I'm having to update my mental model. I thought you were saying you're in equities, they drop, no worries, double your position, you lower your um you know, the rate at which you're going to break even, but you're saying your actual reaction to 2000 2008 was to eject out of equities for a very extended period of time. >> Yeah, cuz I was lucky enough to be able to make money in other things, right? Cuz I'm a a a macro investor. But that's the psychological scarring that you do. And if I'd have just held on to the S&P 500, which is not the most exciting thing in the world, since 1997, it would have made an preposterous amount of money. And this was the same lesson I learned with Bitcoin if you remember. Told you that story where I sold out and blah blah blah and then bought back in and I've realized why is Jeff Bezos so rich? Why is Warren Buffett so rich? Why is all of this stuff so? It's because they basically have long-term holds and they buy when it's cheap. And you started with Buffett here and Buffett's wildly underperformed a lot of people, lot of stocks over time, but really all he's doing he just holds on to stocks for a very long time and if they get cheap, he buys some more. This is exactly what I'm saying. But because I quite like racier things that go up faster because it's not other people's capital, it's my capital. >> Mhm. >> That's That's different. >> If we were going to run an average Joe playbook right now, I think I've heard you just say it, which is okay, listen, if I were going to redo 2000 to now, instead of dipping out and just being gone forever, I'm not going in on debt, I'm going to have a broad-based position, I'm going to think long-term, when things dip down, then I need to get back in, not necessarily double my position, but I need to get back in, stay in as long as whatever thesis I believe about the technology, which you haven't said this here, but you've very much on record as saying, look, tech is the only thing that really grows. So, presumably that would be the the statement that the average Joe, you've got to be in technology in a smart way, meaning no debt, broad-based, and then hold for a long period of time. Is that the average Joe playbook? >> Yeah, and then you think, okay, how much does it psychologically affect me if it goes down? And people tell you they're okay with a 50% drawdown, but they're not. Um but technology stocks cuz they perform so much and crypto because it performs so much over time has these drawdowns. So, you have to be honest with yourself, look yourself in the mirror and say 50% drawdown. So, then say, "Okay, my whole investment pie is $1,000. If it went down 50%, how would I feel? Sick. Okay. Then, all you do is say, "Okay, I'll take that portion out in cash. So, I've reduced my position size. Now, how do I feel if my $600 falls 50%? I've lost $300. I don't mind anymore. That's your comfort zone. That cash is what you use when the market falls. >> Mhm. >> That's what Buffett is doing. That's all it is. That's the whole game of long-term investing. It's so easy, it's ridiculous, but we complicate it. We want to go on Excel all day and look for the latest stock pick, and then we want to worry about, you know, where the latest crash is going to come from. We debate it and do it. You don't actually need to do any of that stuff. >> Okay, I was going to ask you if there was a metric that if it moved in the wrong direction, you would consider AI to be in trouble, bearish territory, whatever, but it sounds like you don't really think about that. >> Of course, I do, cuz I have, you know, I write research and I have to think about other people, and we, you know, just because I have said to you with conviction, "These are my views." You have to hold them loosely because, you know, nobody knows. Things change. Things happen. So, I've actually built myself, I'm going to boast now, with Claude code, an enormous dash not dashboard, it's a platform for Global Macro Investor, my research business, and then Real Vision as well has versions of these, but I've built ones around just this. Supercycle, is it going to be a supercycle? What is the What is the intelligence cycle? Where is the debt risk? Where is this? So, I've got like 15 indications in that that I put together as one single indicator, and that's everything from credit spreads on Oracle and Coe even the other listed ones to the rate of change of the earnings of the hyperscalers through sort of you know the rate of change of the adoption of the technology through to you know there's a whole bunch of these things that make sense. You want to see is that stable? It's of course it's going to go up and down or something happening there that's not good. Then we go back to the so that's the exponential age stuff. Then we go back to the basics is is liquidity going up or not? Are financial conditions amenable for this to continue? Because to your point if there is not enough liquidity for the needs of this beast then we run into problems. Because there's not enough money around. Now what the US has done Kevin Warsh has done this um well he he was part of it Scott Bessent was part of it and Stephen Miller was part of it. They changed the banking laws. And Japan has done the same thing as well. What they're doing is from 2008 until last year basically all liquidity was provided by the Treasury or the Federal Reserve. By the balance sheets and the Treasury general account and the repo markets and all this stuff centrally controlled. What's happened is they're changing the banking regulations to make sure the banks do it. So the banks only two places can two people can create liquidity. The central bank it's uh can create money the central bank or the banking system itself. And they're basically asking the banking system to do it. Japan did this they steepened the yield curve they released the currency valve and what's happened is everyone's like oh my god Japan's going to blow up. The opposite happened. Japanese banks started lending. Why are they doing this? Because it's exactly your point. There is no way the governments can provide the liquidity for this. So, it has to come from the banking system, the lending markets, the capital markets, the savings pools, all of that. And in the end, maybe you're right. In 7 years time, 5 years time, will that be a misallocation of capital? Doesn't look like it cuz this is the fastest growth of anything we've seen. >> Mhm. >> But who knows where that ends up and who does the stupid things cuz there will be stupid things. We're seeing margin debt uh on US equities as a percentage of market cap being high already. And that's like, you know, you don't like to see that cuz it tells you that humans are speculating on all of this stuff. >> Mhm. >> You know, you get that stuff. So, but again, is it likely to end? I mean, again, I've been through the '90s and this ain't nothing yet. It was insane squared. I mean, I I remember being on holiday, I can't remember where it was, somewhere in the Caribbean, some fancy place. Like, no, it wasn't the Caribbean. I can't remember where it was. Somewhere in the world, some very fancy dive resort, Maldives or somewhere. And I was chatting to this American guy, I'm like, "What are you doing here?" He was a guy who climbed the telegraph poles and fixed the the AT&T wires. And he's like, "You know, I've just I was talking about his new $200,000 kitchen he just put into his house or whatever the $100,000 whatever the number was, it was a stupid number. Blah blah blah." And I'm like, "What do you do?" And he was the And I'm like, "How did you do this?" He goes, "I'm a day trader. It's where I make all my money." And it's at that point when you We haven't got there yet. >> Yeah. >> We're not We're just not there yet. >> We'll get right back to the show, but first, let's talk about an assumption that virtually all founders make. You think your company is secure. You've got people to handle it, but nobody has necessarily verified it, and that's where your strategy breaks down. That's just hope. Horizon 3 ends the guessing game. Node Zero is an autonomous AI hacker that actually attacks your own environment the way a real attacker would, your entire attack surface, external to cloud to identity in a single test. It runs 225,000 production safe tests with zero downtime. See exactly what an attacker would find in your environment before they find it for you. Go to horizon3.ai/impacttheory and request your free node zero demo. Visit horizon3.ai/impacttheory. Again, that's horizon3.ai/impacttheory. No commitment is required. Results in hours, not weeks. Now, let's get back to the show. >> What lesson should we be taking right now is where crypto's at in this moment, which I'll say is chopping sideways. Is that a liquidity story? AI's just sucked all the enthusiasm and all the money or is it a fundamentally flawed tech that nobody's going to remember? >> No, I mean, look, it's exactly exactly as you said. There's a certain amount of liquidity. And the reason being is the US is not making a lot of it because what they want is the banking system to do it. But the yield curve is because of the Iran war and slightly pernicious inflation and the oil price and stuff like that, the yield curve's not been steep, so banks aren't kind of wanting to lend at the long end. They're not making enough turn, enough spread. So, that's stopped the banks using their balance sheets enough yet. So, wash, they're they're trying to get that resolved. That's That those are the levers they want. Um and then the banking system should step in, but generally speaking, US liquidity growth has not been big. And that's been the reason. So, if it's not growing super fast, my general um broad liquidity measure is about 4%, but we need to grow it about 8% to fund the debts and the interest payments. So therefore it is enough to fund the AI bet but it's not enough to fund everything else. >> Mhm. >> And they have to do something about it cuz the interest payments on the debt they've got a lot of debt to roll this year. So >> Yeah and they keep jamming at the short end which is liquidity. So yeah liquidity is not enough yet versus what we need for this. They're hoping the banking system takes the difference. Then crypto takes off. I think it all resolves itself. Again they're not They're not stupid. I always take I've known Scott Bessent for 30 years. I first went into his office when he was at Soros and I was at Goldman Sachs in London. He's been to my house in Spain. He was a subscriber to my research for many years. >> Let's go. >> So he's not a stupid guy. These guys know what they're doing. Now do they get it right? Different matter but they know the game. Don't forget we've got a macro hedge fund manager running the US Treasury. Never happened before. And we know we know we we understand his language. He understands our language. He knows the game. >> Yeah no doubt Bessent is an interesting character to me. I think he's becoming more political as the days go by but yeah it definitely feels like you've got somebody that understands how to pull the levers in the right way. So as you look at where crypto goes do you think that people are going to return to it because there is something fundamental there? Like when I look at the Clarity Act and I see Japan moving on there as I see Russia moving on there as you know I get how it creates appetite for debt. You get the stable coins back at one to one. The US started the hype train on that, but we haven't been able to get it across the finish line. Do you think it's That finally goes across the finish line, we've got a fundamental reason why people come back? Um or is it just No, no, no. Crypto is when there's enough liquidity slashing around, people go way out on the risk curve. Crypto's a fun thing that they can gamble on. >> Um all of the above. I mean, you know, we can't pretend that it's not the most liquidity sensitive asset on Earth. Reason being, it's the scarcity in Bitcoin, and then Bitcoin drives further speculation, it goes out the risk curve. But it's also a technology. And that's the important part to understand. Let's use Ethereum cuz it's easy, or Solana. These are just technologies. They're technologies that allow you to do certain things. And what we're seeing, finally, is the entire financial system scrambling to adopt it. So, it's not just about buying Ethereum. In fact, they don't really care about that stuff. Sure, some investors do. What they care about is using blockchain. And Ethereum is the cost of the block space. It's a currency in that respect. It's a claim on the block space. And so, we're seeing the rise of stable coins. It's been huge how fast they're growing. And everybody's built from Stripe to Shopify to Circle to I mean, you name it. Everybody's building out stable coins. So, they've made the decision. Scott Bessent has said, "Well, this is where the liquidity for our debt's going to come from." We want to get to a trillion, two trillion, three trillion of stable coins. Why? Cuz it means that US dominance of currency goes to the end person in the rice field in the Philippines, where they're holding dollars in stable coins, to the fact that three trillion of stable coins is a three trillion buyer of short-term debt. >> Mhm. >> Great. That solves a bit a lot of problems. So, stable coins is good for everybody, which is why it got through its regulations super fast. It's an economic weapon. It's an economic superpower. Increases velocity of money. Does all the things you want it to do. Then the banks are building rails for tokenizing assets. Funds, real estate, whatever it is, because it's an efficient way of doing it. Great. Then the DTCC who custodies and clears all of this stuff does like four quadrillion trades a year or whatever. Well, they're building on this now. The Nasdaq's building on it. So, the use of the technology is going like this. But, they're also relatively early stage. So, the speculators are the difference. So, the the the the building is all happening. The speculators are the difference. Right now, the building is still relatively early apart from stablecoins which are growing fast. A real world assets are growing fast, but they're too small, and the speculators aren't around cuz there's not enough money around. So, you might as well if you've got any speculating in these, you trade. >> Mhm. >> Probability of the of the bull market returning, continuing the secular trend that it's been in since it came out, you know, for me is extremely high because every single bank and every single financial institution and every single government is telling us going to. They wouldn't be regulating if it's not going to happen. The entire banking system is telling us going to happen. Everybody from Larry Fink to Jamie Dimon is telling us going to happen. And yet people are saying well, it's not going to happen. So, that that's highly unlikely. So, it's the liquidity story more than anything that brings in excess capital and excess savings, which allows us to speculate more, which is why they always go hand in hand. Um and I think we haven't got to that point in markets because liquidity has not been growing enough. And my bet is it's coming. >> Mhm. All right, I've got a crazy idea that I'm not even sure how much I believe in this, but it uh certainly is in the back of my mind that one thing that's going to happen is you've got the central banks buying up gold trying to get out from under the US system, I think. And China obviously being the most aggressive among them. I think China's really making a play to I think they're going to try to back the Yuan with gold and say, "No, listen, you guys have us all wrong. Like we're the sensible people." But at a minimum, they're going to be able to get out from under gold. They've stopped their own people from trading paper gold. I think they want to get a Hoover to get all the gold that they can to leave the US, the West, London certainly, and get over into China both in the central banks and with personal buying. And then if they ever need to clamp that valve off, they can do what the US did and revalue the Yuan or whatever they decide to do. But what's interesting to me in terms of Bitcoin is you've got this major play where the number one central bank asset is now gold over US debt. And I think for a digital native, we have call it one generation right now that were kids when Bitcoin came out. So for them, Bitcoin's just a thing. You're going to raise another generation for whom not only Bitcoin will be that. Bitcoin will be like the graybeards. That's what they're into. But stablecoins will just be the native way that they think about it. And as the world begins to say, "We need to get out from under US debt as a reserve asset. We're going to gold." That the digital natives will be like, "Oh, yeah, cool. I want the same idea. I just want mine in Bitcoin because I don't like the idea that it's physical. I have to store it somewhere." And I could see that cuz I don't think right now people actually treat Bitcoin like digital gold. Though for me, that remains the most compelling aspect of Bitcoin. Could you see that psychological shift happening or do you think that's fairy dust? >> Until it gets closer to saturation of its total addressable market um and volatility goes down you're not going to it's harder to substitute because it's volatile. So, if you're putting it in your savings assets, it's quite hard in gold. I mean, gold can move a reasonable amount. You know, 20% correction is not unusual and they can last years. Uh and in Bitcoin, it's 50% 60% whatever the number is, you pick your number. Um so, until it gets deeper more mature um it's unlikely to do that, but that process is the process of getting mature as more people use it, and more people adopt it, and more people lock it away, and less people trade it, and it becomes less speculative, all of that. So, yeah, I think that's coming. Um I Yes. Many of the world's central banks want to be less beholden to the US. Um but the there's no way they can operate without them and they're not going to the system cannot operate without dollars. Um and I've heard from my entire career uh China or choose your own country is going to go and back to the gold Nobody's going to go back to the gold standard. It's never going to happen. It's far too difficult. >> they're moving so hard then to buying gold? >> Well, because their currency's been going down. I mean, think of a central bank, right? Most central banks just do two trades. When their currency's going down they buy gold. When their currency or the US dollar starts going down they actually they buy dollars and treasuries. They have They run this almost all of them run this whole strategy um as a function of their inflows and outflows and and exports because when the US dollar is weak, they're selling more products. When the US dollar is weak, the economy's working better, This currencies are strong. They're making money. They start investing in US assets. And the other way with gold. So, yeah, over time they want to diversify assets for sure. But I don't think anybody's got an answer to how do we get away from the dollar. >> I don't want to put words in your mouth. For me, it's just a question of timeline. Um when I look at what the US is going to have to do, they're going to keep inflating. And if I am in a cold war with the US, which China is in my estimation, I certainly don't want to be beholden to them. I want to get as much strength as I can. I'm already choking them out from a rare earth metal and processing standpoint, from a manufacturing standpoint. Basically, all drone roads lead through Beijing. So, if I can also weaken their influence over me because I hold their debt, I will, which they obviously are selling the debt. Um so, while I get it right now, the system certainly operates on USD and it's certainly not going to change like >> Tom Marlo has said this. Luke Grommen I've heard it from, you know, all of these people. This argument has been going on my entire career. >> Okay, let me ask I'll ask something. >> Not once has it been true, Tom. Not once. >> Well, So, I'm going to make a case that the it's because you're looking at the time scale incorrectly. And that these guys have called the thing that actually is happening. And so, it's just that yeah, it might take another 20 years or whatever, but from a direction of travel perspective, uh it seems pretty self-evident that China is is going to free itself from being beholden to the US. Uh that the US has done itself so dirty from a petrodollar standpoint that you're going to see a weakening of the petrodollar. You've already got Scott Bessent saying we need a Bretton Woods 2.0. So, he admits the petrodollar's not working anymore. So, it's like, well, if we're in this state of flux, I think these guys have been pointing to something that is I'll say it's an inevitability. Every empire that's ever had a reserve currency has lost it on roughly the timeline that we're on now. And yeah, it may take a very long time. So, these guys might have started calling a 50-year um move at year 50. And so, now it's going to take a very long time for that actually to play out. >> what useful information is that? How use How useful is it to say in 50 years' time the world may have adopted a different currency based around electricity and some of them might use gold. How does that help anybody? Apart from create a long doomy story about the world's so [ __ ] the US has lost its empire, the dollar is over, it's all going to collapse. How does that help anybody? This is why I've got real issue with this story cuz I've heard it my whole life and it's it's actually lost people more money than any single other mimetic that's ever been spread. Even more than >> Interesting. >> buy these stocks at the top. Anything. This has cost more opportunity cost than anything I've ever seen. Meanwhile, the dollar keeps going up and they all say, "Yeah, but it's going to crash." >> Well, that's interesting. So, I'm not saying it's going to crash. So, I'm painting a I have a different >> How can they move away from the Go on. >> So, the way that I think about this, the reason I think that it makes sense to pay attention to that move is um I come at this from the perspective of um don't think that you see the future clearly, so don't make like these really extreme bets. Um given that I see people moving away from uh the dollar, I think, okay, central banks are more interested now in holding gold. So, while I would never want a meaningful percentage of my portfolio to be in gold, it's like, oh, cool, that's something for me to put a little bit more money in. Um, so, that's certainly part of my strategy is in this moment, um, I've made a lot of money off of, call it, the straightforward trade, being technology, expose yourself to equities. Uh, it's it's been great. Now, I'm starting to take some of those wins and I'm starting to diversify more, a little bit into gold is one of the plays that I'm making. Um, going more broad into >> I'm more worried, Tom, not about that, which is the sensible way of owning gold. Everyone should own a bit of gold. It's the strategy of telling people that this is all going to unravel in this big world world order change. And what happens is it psychologically affects people that cost them a fortune. >> Yeah, no, I think you got your finger on something. I want to make sure that you you realize I'm not saying that. So, >> I know that the story you're telling is and people won't hear what you've just said, which is like, oh, I've got a percentage of my allocation in gold because it's got a place in a portfolio and it feels right. I can see a trend there, which is very rational, logical, to the US is doing this and this is a 4D jigsaw puzzle of which the Chinese are going to win cuz they're going to take over the the rare earth metals in Africa and the Belt and Road and the US can't see this and you know, and then suddenly there's going to be a new currency and it's gold, but that whole stuff gets people into a feeling that it's all about to end. Any day now, my position in the US or the US economy or US equities, can't you see they're going to take it away from us? >> Well, let me let me ask you a very direct question. They're all selling. Well, yeah, they're all selling clicks and certainty and all that, totally understand. But but let me ask you a very specific question. Because I have this belief that oh, this is something that I think it's happening in the background, I don't own any long-term US debt. So, I own a ton of US debt, but it's all short-term. So, that's a very direct way that that belief plays out in my portfolio. Do you think that's crazy? And you own long-term debt and you're like, "Yay." Or >> No, I I I was a bond holder for years cuz it was a bond bull market. As we went, you know, as the baby boomers went through retirement, it drove down um inflation. I still think I don't think we're going to have long inflation. I think we'll have deflation back again cuz that's what population shrinkage and technology does. But, you might as well back the innovation, which is owning the equity than owning the bond, which is the US government's liability. So, no, I wouldn't I wouldn't own them. Short-term debt, great. Interest rates are pretty decent right now. Um you know, gives you a bit of cash balance. Not a problem. If the world, you know, if the markets correct sharply, you've got some cash in short-term T-bills. Fantastic. You can use them. They're all super sensible, you know, to have a bit of gold, a pot of cash earning some some some yield, um some technology equities, bit of crypto. It's a very sensible portfolio. >> We'll get right back to the show in a second, but first, I want to talk about the customers you're losing without ever knowing it and why today's episode is brought to you by Qual the business phone system built so you never miss an opportunity. Most businesses track leads coming in. It's very important, but very few track the ones that slip through the cracks after a dropped follow-up or a buried voicemail. Qual fixes that. All your calls, texts, and voicemails now live in one place so your team has the full history and nothing slips through the cracks. Qualtrics is the number one rated business phone system on G2 and trusted by over 90,000 businesses. Qualtrics built-in AI agent handles after-hours calls, answers questions, and books appointments so you never miss a lead. When money is on the line, always say hello with Qualtrics. Try Qualtrics for free plus get 20% off your first 6 months at Qualtrics.com/impact. That's q u o.com/impact. There's a moment every entrepreneur remembers for the rest of their lives. The moment a stranger, someone who doesn't know them, doesn't owe them anything, buys their product for the first time. That's the moment that business stops being a dream and starts being real. Shopify is what gets you there. From day one, everything you need is already built. Professional templates and AI tools that make your storefront look incredible instantly. There's no coding, no designer required. You get Shopify checkout, the best converting checkout on the planet, so customers actually follow through with their purchase, and Sidekick, which is Shopify's built-in AI assistant. It's ready to answer every question that comes up so nothing slows you down from getting your sales. We used Shopify when Impact Theory had a merch store. Everything was so easy from day one. My team loved how simple it was, and honestly, I was jealous of entrepreneurs who had that leg up that I didn't have when I first started out. So, start your free trial today at shopify.com/impact. Now, let's get back to the show. You mentioned earlier the K-shaped economy, it's going to keep going. Oh my god, this is It's so funny talking to you is like a fascinating mirror on myself. So, I look at the K-shaped economy and let's say the let's say Say doom now, come on. Yes. So, well, so I actually don't buy into doom because that would assume that I can see the future clearly, which I know I cannot. But, um when I >> You assume it's going to rain tomorrow. >> No, but I own an umbrella and I keep it by the door. That's probably the right way to think about me. Uh so that should I go out and see raindrops on the street, I'm not like where the [ __ ] do I get an umbrella. Um but so I look at the K-shaped economy. This is probably the thing that I think the most about. This is the only reason that I started researching the economy in the first place was um I I at the point that 2020 hit, I was still only a few years out from having thousands of employees. A thousand of them had grown up very hard. And so they're, you know, at the the wrong side of the K. They're on the wrong side of the K. So, that's what got me engaged in all of this. Now, I see in the US the rise of the DSA, which occupies a I am emotionally triggered by the DSA, so that already gets me invested in all of this. Um but do you see any um concerns about doing economic things that will ensure that the K remains a K or to you is that a nothingburger and it will peter out on its own? >> Which, the K-shaped economy? >> Yeah, cuz I I see growing tensions in the US that will not resolve until the K is closer to the middle. >> I say it's going to get worse. And so is the >> And not not a big deal or it's going to get worse in this country. >> deal, but it's been happening for quite a while, which is why the US has done this in politics. And the UK's done this and everyone has done this, right? There's virtually centrism has disappeared from everywhere. Um why? Because one group want to blame one group and one want to blame another, but as I said, a lot of this is down to demographics. Blame World War II if you want to blame anybody. Um and so this whole issue doesn't go away. And that fight has also been the fight between capital and labor. That's always been the balance of politics and economies. We're about to get something bigger, which is we're going to split between the accelerationists and the decelerationists. Those who want to embrace the technology to give them the superpowers of economic superpowers or physical superpowers or mental superpowers versus the people who are going to go suck their teeth and go "No, you don't want to do that. This is dangerous. This is terrifying. You don't know what it's going to do. We're going to lose our jobs." Right? That is going to be the defining battle. And then it will eventually morph into who wants to give the AI robots rights, economic rights. Uh that's all coming. So, we're going to be divided for a long time. The issue is the printing of currency to service the debts, to keep everybody's livelihoods basically okay. Costs the poor side of society everything. And I don't know the answer cuz every side has tried it. They all see it. If you're If you're in the US, if you're a Republican, the center of the country, screwed because of the K-shaped economy. If you're a Democrat, city dwellers who aren't rich, screwed because of this. There's their voters. Both have tried, both have been in power, none of Nobody's resolved any of this because it's They can't. The answer would be stop printing currency. Mhm. What does that do? That collapses all the pension system. Cuz the equity market. Which way do you want it? Do you want a 5% You want to You want to 8% debasement, which is the cost of the put? Or do you want an 80% haircut? You choose. That's That That is That is the game, unfortunately. It's horrific. Most people choose the 8% cost. >> Mhm. When you look at um as a macro investor, and you're looking at historicals, uh do you go okay, this all leads to Maggie Thatcher or what do you think this leads to? Is this French Revolution or Maggie Thatcher? >> The fourth turning. What who do you think is going to be running economies in 10 years time? It's obviously going to be AI. The decision-making changes wildly. We're already seeing AI in every company. It's not fully on the board of many, but it is on some. If I go to the Middle East, which I go often, and go and see the sovereign wealth funds, they've got AI on their investment committees as a person who votes. >> Wow. >> Yeah. And they've got it in companies and they've got it in government. We all will cuz it's smarter and better at decision-making, more level-headed, and can see the picture better than we can. So in terms of optimal allocation of resources, which is what this whole game of humanity does, it's better than us. We're all using it for marketing strategies or whatever it is. Why? Cuz it's better. It's faster, makes better decisions. It sees with more depth and connects more dots than most people do. It's not as good as the greatest people on earth, but it's bloody good at everything. In fact, every single topic known to humanity and every single mental skill known to humanity, it's up there with the point zero zero one percent. So of course it's going to be running governments. So when we talk about this in the future, do you think that Gen Z are going to say, you know what, we should, you know, have the the the Gen Xers or the whatever? No, they're all going to say, of course it makes more sense to use AI. So, that's where I think it's going. And that changes now. Is does AI come good, AI come bad? You know, that's the whole world of sci-fi and we're going to find out. That's where we're going. We'll find out is it good or bad? And my guess is it was built on our intelligence, it'll be both good and bad like we are. But generally speaking, it'll just keep moving forward in a way that doesn't blow everything up. Cuz if it removes us, it removes the compute that makes it. The messy compute. Humans have this very weird compute because we do all sorts of random things driven by qualia. So, we create this very complex compute. It doesn't do that. Machines don't do that yet. What do you think Neuralink is about? It's the it's the direct connection between machine compute and human compute. Okay, that's super interesting. It's unlikely to get rid of us because it can't do without us like we can't get rid of the trees that we talked about before, the ecosystem of intelligence that allowed for humanity. If we did that, if we destroyed the ecosystem, the planet, nature, we die. We're all built on each other. >> anybody make that argument. Um tell me more. It is not immediately uh obvious to me that that would be true. >> So, we were talking before and it's a longer conversation for another day, but I built a a thesis which I call the universal code, which is even more grandiose than the everything code. Um and more heuristic, even though I come from the acknowledgement is I don't know I'm no expert in physics, philosophy, mathematics, any of this stuff, but I stumbled into this whole big thesis. Part of the thesis is that intelligence is that is that things atoms, particles, all the component parts eventually start creating coherence and networks. And what it does, and I'm going to go back to firstly humans. So, humans we are hugely curious. We network with each other. We then start compressing all the intelligence of everything around us. That is what science is. People don't think of it that way. Science exists. I mean, these things exist. We think of science as a breakthrough. It's not. It's us figuring something out that already exists. We just didn't know the story, and then we find the story. So, we're compressing all of the knowledge of the things that were around us. How do, you know, how does geology work? How does biology work? How does physics work? How does all of this exists. We're not making it. So, we're compressing the intelligence of everything around us cuz we're the apex intelligence. The AI is just compressing human intelligence, of which more intelligence will come out. And the whole process of the universe is exactly that. It is forming coherence, compress that creates intelligence over time. And that eventually compresses and forms more cuz the universe solves for output of intelligence per unit of energy. So, we So, AI is not going to kill us by definition because we are part of it. Like nature is part of us. And the atom The water is part of the trees. I know this sounds woo-woo. It's not. It's actually just physics and biology. This is exactly how things work. You can't kill it off. Take the water from the planet, all the trees die. Take the trees, all the humans die, and all the animals die. Take the humans away, all AI dies. It's all the same thing. We are the same thing. We're just nodes in a larger consciousness. And a larger computing process. >> How are you investing through what is obviously the biggest technological revolution in human history? >> I have a framework which is that the universe, markets, and everything we do solves for output of intelligence per unit of energy. This is one of the reasons why we've seen technology stocks just outperform everything. And this is why crypto, which is also part of the same idea, outperformed everything over time. So, it's been for me that basis. I think this is a super cycle. Listen, >> the the technology is sort of the ultimate thing that this has all been aimed at, and therefore it can't get big enough. >> It can never get big enough. It just keeps attracting more capital cuz it's the highest use of that capital. Um and it has the most output in terms of returns, but also for humanity. You know, the humanity scale of what is happening right now. You know, people don't realize I mean AI is easily the greatest discovery humanity will ever make because we've created intelligence. We've put electricity through silicon, through sand, and created intelligence. All intelligence in the past was putting electricity through carbon. Now, electricity through sand is a million times faster. So, the silicon substrate is a million times faster than the than the biological substrate. >> Wow. >> Which is why we can't compete, is which why we feel so far behind all the time. And that's creating not Metcalfe's law, but Reed's law, which is Metcalfe's law squared. And it's the fastest adoption of anything, not any technology. Reed's law doesn't even even exist in biology. >> Okay, I knew we would eventually get to to Reed's law. I did not expect it this fast. So, let's lay um a marker for people. So, one, um there's going to be two visions that are going to compete in this interview. So, mine will be paranoia, historically grounded, um nothing ever goes up forever, even if only because humans panic and run for the doors. You have debt mechanisms and things to think about. We'll we'll debate that when we get there. Then you've got the um far more optimistic view that you have, which is always sort of where you and I are as archetypes in this conversation, um that you said a you can AI can never get big enough because this is sort of the ultimate expression of what the universe is attempting to do. So, capital is going to flow there because it is the right place for capital to go. Okay, now you've you're very famous for Metcalfe's law, which is the power of networks. Um so, they grow exponentially or the the um value of them grows exponentially. Walk me through what a double exponential actually means. Like, how do I even wrap my head around that? >> So, if we think of the internet as Metcalfe's law, the value of the internet is basically the number of people using it, the number of connections that they have. And formally, it's like the number of the nodes in the network squared or something like that. But, it doesn't matter. It basically for simple idea is the more people in the network doing valuable things, the more valuable the network is. Makes total sense. Now, when you build another system on top of something already growing at Metcalfe's law, you can get this double exponential. So, to put it in simple terms, AI is basically a compression of every single thing that exists on the internet. And more. You know, if you think about AI, Tesla self-driving is everything it sees in the streets, or Google will have everything it sees on every video that's ever been created on you You know, there's a lot going into this, vast amounts of data. What you're doing now is you're compounding information, running it through a silicon substrate, which is faster than anything we've seen. Um as silicon chips get better and better, and which is why Nvidia is so powerful and all of that stuff. And what you end up with is a double exponential. So, double exponentials are not natural. Or they weren't natural. They've never existed. People talked about it as like Reddit subthreads would be built on top of Reddit, and they were um Reed's law. I don't believe that, but this is provably so. So, provably so is Anthropic's revenue scaling is the fastest of any company in human history by an order of magnitude. Now, they'll do a they'll do something like a hundred billion in revenues this year from 18 months ago of zero. >> So crazy. >> Then we look at OpenAI. So, before OpenAI, another technology built on the internet was cryptocurrencies and blockchains. They scaled faster than the internet itself because they have this incentive mechanism of of the network having actual value that you can invest in. So, that was the fastest adoption of any technology the world's ever seen, and anybody who's ever seen me on your show have heard me say that a lot. Then AI came out, and it got to a hundred million users in a week. And now OpenAI has a billion users, and Gemini has a billion users. I don't know what Meta's got, maybe 300 million users, maybe 500 million users, even even nobody really thinks of Meta really as having cracked it. And then you've got Anthropic and everybody else, and then the Chinese labs. I mean, this is stupid. And we wonder why we feel left behind, we don't understand it, we can't keep on top of it. It's because it's moving faster than we can comprehend. We have no way. Exponentials are hard enough. Double exponentials are impossible. >> From a philosophical view of looking at why AI is ultimately going to be the substrate through which we engage with our lives, whether that's a smart refrigerator, a smart blender, having it help me with my homework, whatever. That you've answered that, I think, very, very clear. Where my paranoia comes in, and the reason that I look at Buffett as he's backing away, I don't not exiting entirely, but certainly backing away saying, I don't think that people are investing in a sensible way, that resonates with me because I look at something like >> Although, he just did add to his Google position. >> Yes, but he also just backed out of his Apple position. He's sitting on more what people refer to as cash, you know, historic levels. And now, for me, I just look very simply at the Cape ratio. And when I see the average has been 16 to 17 X, and we're now sitting at 40 X, meaning the revenues that are actually generated by, let's just round it to the S&P 500, versus what people value the S&P 500 at, have become not a-historic, but we're reaching historic highs that we haven't seen since the dot-com bubble burst. And I go, okay, I believe the technology, I believe everything that you've just said in terms of its importance, its inevitability, all true. >> Is it priced right? >> Is it priced right? And right now, we've already seen OpenAI plateau. So, if you were just all in on Open AI, you're probably going to get hurt unless you start adjusting. The same thing could happen to Anthropic because yes, its revenues are growing at an insane rate, but it's still burning capital faster than it's generating the revenue. And so I go, "Hey, despite the inevitability, there are going to be people that get hurt economically as we repeat the historical pattern of every revolutionary technology ends up taking on debt to the point where it wipes out the first round of investors, and then it's the inheritance generation of investors that actually make the money." >> So, there's a number of things within that. Firstly, the Cape ratio, just don't forget debasement. Debasement is what moves E moves the price. So, the PE ratio, price goes up because of the debasement of currency, and earnings grows with GDP or, you know, in the case of Anthropic, different, right? So, that's always going to trend higher because we keep debasing the currency. So, you need to kind of think of that in your head and adjust it cuz there's an >> So, before you move on from that, so we've got if that's true cuz the Cape ratio is looking at a 10-year average. So, are you saying that the moment that we're in now is so different than the 10-year rolling average that it's completely expected that we would no longer be in the 16 to 17 range that we have been for the last 100 years or so, and now 40 is like the new normal? >> Yeah, so what we have seen since 2008 is a use, yeah, a printing of currency in a very obvious way. And the printing of currency has changed valuations because the price keeps rising, equities keep going up, but the earnings are more constant because that's not affected by debasement. That that mechanism is part of what makes equities look expensive. Ignoring all of that, to go to your point is, you know, is it dangerous now or when will it be dangerous? I mean, all investments are dangerous. They always get the down cycle. You always get the up cycle. Having lived through '95 to 2000, we ain't seen nothing yet. Nothing. Um so I don't think in the basis of collective insanity, we're there yet. >> Okay, so sorry, just to really restate that point. So, uh hey Tom, I've lived through more investing cycles than you, which is obviously true. I I am so recent to this game. So, um I've seen way worse. We're not near the sort of um just completely detached, nonsensical investing yet, but you also believe that yeah, there's a likelihood that some people who are doing something foolish will get hurt. Is that true so far? >> If that's not true, then there is no risk in financial markets and you know, of course it's true. It's always been true. The job that we have is where are we? Where are we in that cycle, right? >> Two questions in, I want an answer to that. And then what do you do to protect yourself? Is it No, I'm just so good at knowing where we're at, I never miss. Is it I do my best to know where we're at, but I have some other downside protection. How do you How are you thinking about that right now in this moment? >> See, I'm an old fish in all these things. Like I'm with crypto, it's the same thing. I look at the long-term secular trend and understand that markets go up and down within it, and it doesn't phase me. In fact, what I look for is when you do have a correction in a long-term secular trend, you buy more. Just look at the charts of the Nasdaq since 1982, since the advent of the um computer, the desktop computer, it has just been this. And everyone goes sucks their teeth and goes, "Well, 2000." I mean, [ __ ] it was just a blip on the chart. The answer to 2001, when it fell, I don't know what the Nasdaq fell. Then 50, 60%? Was to buy more. It's always been the answer. We've had this discussion about crypto and Bitcoin and everything else. People forget it every time because emotion gets in the way of what you're trying to achieve. Now, you know, those losses, it depends whether you can take the volatility. And if you can't, then invest less or worry all day about the bearish story and try and use your charts and try and figure out the risk parameters and all of that. It's much easier not to do any of that. Just invest less. And then have some money held back for when it does fall and then go, "Amazing. I can't you know, pop open a bottle of champagne, buy some stocks and you think, great, I've got a discount." But people can't do that. But that's what I've taught myself to do is you know, be very happy like seeing where crypto has been against its long-term trend, let's say Bitcoin, it's on that trend at two standard deviations oversold, which is historically every time a fantastic time to buy. It's when nobody else wants to buy. And so So, that's how I That's how I I think about that. But the point being is there are ways of measuring this. Nothing is perfect. So, you you're looking at, okay, so the debt is an issue. So, then we break apart the debt and look at it and say, "Okay, how indebted are these hyperscalers?" I think the technical answer would be [ __ ] all. >> You're saying not at all? >> 4% of their market cap in debt. It's fuckle. How much cash do they generate more than God? I mean, these things generate huge amounts of cash. So, they've gone cash flow negative. Oh my God, it's the end of the world. The largest cash flow generating machines on Earth have gone cash flow negative to build what? Have they wasted it on wine and women and what drugs? No. They've built enormous amounts of compute. And they've got the revenues to pay almost any interest rate on the debt ad infinitum cuz they're 4% of their market cap in debt and a fraction of their cash flows. Now, there's a bunch of others. There's CoreWeave, there's maybe Oracle, there's a bunch There's a bunch of special purpose vehicles. Could those blow up? Yes. So, let's run through an even bigger scenario that somebody like you will weigh on their mind. What happens if OpenAI screw it all up? They've got trillions of dollars of obligations and it all goes wrong. What happens tomorrow? Firstly, the US government steps in. There is no way that they can allow the US to slow down on the race for intelligence because you've got another nation also on the race for intelligence and anybody who wins this wins everything. And so, in game theory, you can have no winner. And everybody has to go max all out. But more importantly, it's the same with the companies. The moment OpenAI goes bust, let's say, all of the compute gets bought in seconds by the fastest bidder. But the US government can't allow that either cuz imagine Google buys all of the compute or whatever it is, all of the compute licenses, everything. If Google buys it all, then Google becomes the most powerful company on Earth and they have the monopoly cuz if you double compute, it's proven, you're basically accelerating exponentially the rate of intelligence output. So, it's actually very hard to have massive economic damage. Yes, you'll have a shock, but you don't have anything long-lasting. It's the same happened with the fiber optics. You know, they all they all got written off, but the debt levels were so much higher back in 2001-2 in those companies. They were insanely leveraged, which is not there yet. >> So, do you remember what the rough number was? >> I cannot remember right now, but they were just they had, you know, huge amounts of um negative free cash flow. In fact, they made no money. They had no cash flows. Don't forget, they had zero cash flows really at that point and it was all CapEx. Okay, that's very different. That was the building of the rail roads was all CapEx and no trains running. It's only when the trains were running did you make money, but you had to lay all the tracks and everything else. This is very, very, very different uh by who's doing it. But again, will you have accidents on the way? Of course, you have to because capital must allocate to the most efficient path and people who get it wrong will get punished by markets, people who make mistakes, absolutely. So then, is this affected by the business cycle? Classically, the business cycle would peter out as liquidity starts getting withdrawn and the Fed raise rates. Firstly, we know from 2000 2022, these guys don't really care about the Fed raising rates because they have more cash than God and they sit on huge piles of the stuff and it earns them interest. It doesn't matter if the interest rate goes up a few percent, doesn't really matter. Does matter to the rest of the economy, but not into this race. But also, more importantly, the CapEx is the largest build out in all human history that's happening now. So again, we live in extraordinary times that I keep using this the phrase in all human history and they're all real and they're happening now. Talk about simulation. I mean, it's ridiculous. So this is the largest cap capex event ever and we barely started. Don't forget this entire thing kicked off 3 years ago. We haven't even built the data centers. Sure, Elon can manage to get a data center up and running in about 4 or 5 months or whatever it was, which was preposterous. But we're only at I mean, the capex plans go out to about 2030. That's what the numbers we're seeing that that the numbers of spending. But I think we're about 30% of where we should be on the builds this year because the electricity permitting is so slow. There's There's factors that are throttling this build-out that makes it quite hard to build excess capacity. Um So they're struggling trying to get as much of this build-out done as possible. Uh and so I don't really think that's an issue, but the capex then comes online. So every time you put capex online, you increase your compute, which increases you in the race, which allows you to monetize it more. So this dip in in uh free cash flow is really an investment in 2 years' time. >> Okay, so one story that I hear is that um there aren't as many companies clamoring for the compute uh as they expected there to be. You've got somebody like Elon who builds out that colossal data center only to then rent that um compute back out because he can't justify with his own AI company. And so people are like, "Hey, this is really um about two companies essentially eating up the compute that we have now." And whatever story is being spun that says we have this infinite need for compute just isn't born out by the fact um that you have somebody like Elon who's just now renting it back out. He's I think he had a third company in there that I didn't recognize the name of. Um but it it isn't like there's 300 companies that are going after the compute. This really is a story right now of two dominant players. >> Well, cuz it's so expensive. This is not an easy game to win or to start. But Elon got there and he's done pretty well with xAI. He had to build the compute. He had to build those data centers. But now he's built a bigger and better one, which is why he can rent out Colossus. Meta have so much data cuz they've got this massive network of information that they hold anyway that they have compute. So, why not lease out your compute for huge returns, 30% 35% margins plus, and earn money on that to fund your build out of the next phase of this. So, we're seeing that Google are doing both sides of this equation as well. They're building massive amounts of compute, using massive amounts of compute, and um leasing massive amounts of compute out to others. Amazon doing the same. So, everybody is using all of this at full capacity. Everything is at full capacity. You look at the price of every single GPU, they're all going up. Rental rates of GPUs, all of them are going up. So, what you've got here is we're in the middle of Jevons paradox. It's the more abundant you make all of this stuff, the more demand there is for it. And the demand curve has been insane. It's insane the demand curve. And think about it. What is the demand for intelligence? It's infinite. Cuz intelligence itself with the rise of agents has its own demand for intelligence. Cuz intelligence is the only thing that keeps compounding over time. So, all I'm saying is this is the fastest um the secular trend of Reed's Law, you will have some bad allocation of capital. You'll have people screwing it up. Um you will have periods of time where the markets get nervous about all of this, but this is not going to be stopped, not in the greatest race that's ever been, which is the race for intelligence. >> If some of the debt starts collapsing and we start having a liquidity problem because lenders start going, "Whoa, I don't want any more exposure." Um it feels to me like this could get fragile pretty fast. And so, if that's true, then the undergirding thing becomes, "Yeah, Tom, that doesn't matter because this is US v. China. The US will print money until the cows come home if that's what they have to do to make sure that this industry keeps growing." Is Is that what you see as the ultimate backstop, or is there something else? >> Well, that's one of the backstops. The other backstop is anybody who's got cash who can buy more compute will. Uh people can't get their head around this is a good of infinite demands. We've increased We're about to increase the TAM, the total addressable market of the planet by all the agents we're bringing in, who want compute, who are doing tasks that create compute. So, it isn't unlike anything else we've ever seen. Everything else was defined by human demand. This is not. So, that's change That changes a lot of the equation. The other thing, it is observably wrong that the increase in demand for Chinese models has lowered has made it more fragile for Anthropic or OpenAI. Observably, their margins have remained. So, what we're seeing is the pie growing is not a fixed pie. That makes sense if this is the largest good and Elon will talk about this, too. The largest thing, the largest good you we've ever produced is intelligence. So, therefore, the pie is infinite. So, the pie keeps growing. Now, is is there periods of time where the pie slows down? Yeah, most likely via politics than anything else. Um but the pie keeps growing. And therefore, if if they can maintain margin, will they maintain you know, 55% margins? Maybe not. But all the hyperscaler are going up as well. That's insane that the margins are all going up across every player in this. So, there's nobody losing margin because the demand growth is so big. Now, when you look at the debt, who's got the debt? Well, you know, there's a bunch of Google debt that's now been placed with, you know, pension system and stuff like that. But what is the risk of Google not paying the debt? About as close to zero as possible. At least for now until they get to a much bigger numbers. Um Coreweave debts and a bunch of these data center debts, yeah. They'll nuke some of those in special purpose vehicles that are probably leveraged. That's ugly. What happens to that if somebody buys all the compute cheap and the thing accelerates? So, but so we can have the it can disjoint. So, I'm not saying it can't. But what happens is the speed of which it recovers is going to be really shocking for people because the moment you have distressed assets that produce more intelligence at cheap prices, they snap and intelligence grows faster. More output, um more things we can do with it, the easier it is to use, all of this stuff. >> Okay, it sounds to me like you're betting on the industry or the technology, depending on how you want to say it, versus any one individual player. Is that actually accurate? >> Yeah, correct. Correct. And then your idea is, okay, where do you want to place bets within this? You know, you've got the playing field of bets. Where's risky? Where's not? Where's interesting? Yeah. Interesting to me might not right now might be Oracle. Never fade Larry Ellison is one rule in life. Right? The guy knows what he's doing. You look at the share price of Oracle since he started the company, it is unreal. So, has he has Larry lost his mind? He's gone to see the dead. Or is he acknowledging what I've just said is spend the money as fast as you can now, build out as much computer ahead of everybody else, and you have a massively growing margin business. That's what I think he's doing. The market doesn't see it that way. So, sure. Okay, maybe that's one way of looking at it, which is the argument that if they fall fast enough, you get to some interesting opportunities. You know, then we look at um Amazon, which is a very interesting example. In fact, super rare example, maybe outside of SpaceX and others. Amazon is an example of the productivity miracle that is likely to happen to the economy. So, Amazon is using robots. And the Amazon human uh the Amazon robot human ratio is nearly one for one now. And in the next few years, there'll be more robots than humans working at Amazon. In fact, Amazon's hiring has been flat, and their robot hiring has just kept going up vertically. They've got self-driving vehicles. They'll have self-driving deliveries. They've got drones. They've got every element. They've been way in advance of anybody else. They're running like the greatest logistics company the world has ever seen. >> It's incredible. >> Using the cutting-edge technologies that they built themselves. That's basically what the economy is going to do. And you look at their margins, they hold up really well, and in fact increase over time. Um because they get more efficient and that's what the economy's likely to do which means that for you and I, there's more money around. Becomes more profitable to do things. It's more productive to be a human or to run a business. >> I want to drill into that cuz I think that there's um a base assumption that you have that may not be familiar to a lot of people. My question is, will that artificial intelligence be an economic actor? Will it create surplus? If so, how? And then how does it get the dollars that it needs? Because when I think about right now the way the governments operate, they basically intentionally inflate away innovation to get more liquidity into the system. Do you see the Fed as one example and Warsh has talked about this, but do you see the Fed going, "Okay, AI's going to be deflationary because that intelligence is going to innovate rapidly. There's going to be more stuff for people to buy. Uh therefore it'll have a natural deflationary pressure cuz there'll be more goods chasing the same number of dollars. So I, Kevin Warsh, am going to inject more dollars into the system to basically keep up with that. And so now through some mechanism, QE or whatever, uh the there will just be more dollars floating around in the system. The AI is going to win some of those dollars through its own economic activity and like it just keeps growing that way or do you see something completely different than that? >> No, I mean there's look, there's a lot to unpack and this is where we get into that idea of the economic singularity when the whole economic structure of the world changes. Elon spoke about that with the interview with the economist as well. And it's something I've been writing about for a long time. Is you get to a point where productivity increases and it starts offsetting the debt growth. >> And are you thinking of that as productivity increases per human or productivity increases per agent? Like what how are you running that math? >> That's a That's a great question and that shift is something we'll we'll we'll come on to. So if GDP growth equals population growth plus productivity growth plus debt growth. Right, population is you know if you look at the labor force labor force participation rate, it's declining over time. The rate of population growth is declining in the US and most other Western countries now. So humans are shrinking in the countries that have all of the economic output and all of the wealth. And they're going to get replaced by AI and robots. I mean I employ lots of agents to do stuff for me. And I have to pay for their compute and they in return they do economic activity or activity for me, whatever it may be. Okay, fine. So we're starting to see the I mean that that's only been going for a year, don't forget. This is how fast this shit's moving at. So we're starting to see agents everywhere. Elon's just released a platform for agents, everybody is it's it's literally everywhere. It's all happening and now there is more um bots and agents on the internet than there are humans and that just keeps going exponentially. Huge network effects of these things. And they use computes and they use energy and you know, all of those things and that gives humans jobs building data centers and all of this stuff that we >> can say. Um but also we bring the robots into the equation. So we're increasing synthetic humans here. And what is the And that's productivity because these things are very productive, they don't sleep, they cheap to run, all of this stuff. So what does it do? It creates crazy outcomes for GDP. If GDP is growing at 10% or 20%, which is very possible beyond 2030. It's very possible to see that. It's very hard to debase your currency that you can create inflation. >> That sounds like a challenge. >> It's a challenge, but you don't you don't need to because debt as a percentage of GDP goes down. This is what they did in the 1950s. And you and I have talked about this in the past. So that was the financial repression of the 50s and so it's you don't need as much to debase that so you can give it to the people. And it still won't be inflationary because you can increase let's say from call it 5% right now, 6% and it goes up to 8% but if they can take it up to 10% you're increasing a massive amount of money and as you said goods prices still fall. We can produce so much stuff for nothing. Everyone can produce software for literally zero. $200 a month. It's insane. What we can do. And I built so many things myself. I'm non-technical, never programmed, never looked in my GitHub, not looked a line of code ever in my life and I've built 20 products. All of which I've rolled out to Real Vision members, Global Macro Investor members, things I've used myself. Uh it's extraordinary. So it's now become free. So this you this absolute abundance is what is coming. And we'll see it in medicines, we'll see it in just every part of this intelligence touches because its job is to make everything cheaper, more efficient. That's what it does really well. So I just don't see the bearish case of this. Will it replace jobs? Absolutely. But there's plenty of other roles for humans to do to make money from. And one of the good things about all of this is we're actually losing workers right now because the labor force participation rate is falling. And it's you know, from from my um work it just keeps falling because the birth deaths rate if you're going into the future gives you a, you know, a 60 18-year lead on it. >> Mhm. >> So, we've just got a complete collapse of the labor force participation rate to come. And what we're going to do is keep offsetting that for a while, and that's good. So, the boomers retire and the boomers die and it gets replaced by increased number of robots and stuff for a while. So, it's not the shocking transition that people think. It's a transition we need if not the economy collapses. >> There is a a caveat in your speech that's doing a lot of work, which is that we're going to see GDP growth 2030 plus because right now our GDP sucks ass and we've already brought a ton of um agents on now. So, it sounds like the bet that you're making is that, "Look, this is only 3 years old. For as good as things have gotten now, they're just going to get so much better. So, don't be alarmed that you're not seeing it yet." >> go back to Amazon, Tom. Amazon has been ahead of this whole thing. It's like the US economy in 5 years' time, 10 years' time, five five seven years' time. That's what Amazon is. They use robots more than humans. >> Yep. Okay. >> They are using drones to deliver stuff. They're using self-driving electric vehicles to move around. So, they are an extraordinary example of a company that is gaining productivity. So, it uses less workers for more output. And that chart has just hooked up massively as Amazon started really leaning into the technology side of this. So, that whole hookup in in productivity of Amazon, you know, how much output per human employee has gone through the roof. That is exactly what we're about to go through and Amazon is showing us the way, but people don't want to believe it. >> Um that is the economic engine becomes you have artificial intelligence. It will be an economic actor whether it's because you're able to now have a thousand employees that are all all artificial employees doing a thing or um the AI itself begins to be self-directed and doing its thing. But ultimately, we're going to have to have a way to um get money into the system. >> They don't need money into the system if they can be economic actors. So the first stage is that economic actors for us. So let's say I want to I mean I think Shopify connected it. There's a whole bunch that connected it so you can have your agent go and buy stuff. Mhm. We're seeing this scramble for stable coins, right? Back into crypto land. Massive scramble for this stuff. Uh the guy who built Stripe is like you don't understand all of the transactions are going to be agents. So agents can instantly transact on behalf of you as an economic actor. So when you launch your game, my agent will pay you for its for access to the game. My agent will play the game for me and then I can take over when I'm when I get to the the level I want to do all of this, right? That's all happening. That will absolutely happen without question. So at first this directed economic actors I've got many friends who've given their AI agents autonomy and said here's a hundred dollars come back and make me money. Um and some fail, some some work, but it's a learning process. And we're very early. Again, most of these agents have been around the whole agent ability for us to use an agent about a year old. Yes, in the frontier labs inside they've had agents, but for the public, no, didn't happen. So they're going to be economic actors directed by us then on their own. Because networks over time form coherence and these things are already forming networks. And we will see eventually, and people aren't ready for these kind of conversations, but these things autonomously will do will act and they need to be fed like humans do. What do they eat? They eat electricity and compute. So, they need these things like we eat food and water. You know, one is our life force and the others our energy supply. They have to do the same thing. And so therefore economic payments have to happen cuz if not, they die, obviously. >> Yeah, I get that. So, the part that I'm missing is why you're saying that money won't have to be put into the system. So, here's why despite hearing all of that, I'm still not understanding >> right? Velocity of money goes through the roof because you can make micro instant payments in everything. That's the whole idea of the stable coin rails and the crypto rails. Is what they're allowing is velocity of money so money can get reused multiple times. You actually don't need as much money if it's if there's velocity of money, if it's not an inflationary environment. So, that's one of the key things of this. You don't need to inject money. Money's not a fixed pie. You create money by economic opportunity. >> You you well, sort of. So, you have velocity of money, yep, sure, got it. That's going to create GDP, but that's not going to solve the problem of how are we actually going to inflate the debt away? So, one of the points you made >> need to inflate it away cuz GDP grows fast enough. It becomes too small to matter. >> So, you're saying just by increasing velocity of money >> No, just by increasing growth of the economy. So, let's say >> Well, so hold on. The growth of the economy is a set of variables, one of which is velocity of money. So, which variable, if it isn't velocity of money, are you talking about increasing? >> It's productivity. Productivity is the big one. You move productivity from You move productivity from 2% to 5%, you completely change the game. But productivity can go much higher than that. We just never be able to do it before, but this is a very different situation. And again, we go back to the '50s, after World War II, US debt to GDP was 100 and something percent government debt to GDP. >> Yep. >> By the end of the '60s, it was like 15%? >> Mhm. >> And what they'd done is the economy was growing really strong, and they kept interest rates down, and so what happened it was debt costs and debt compounding ended up growing less than GDP. That's what did that. >> Yeah, but am I crazy? We were basically pulling money out of the rest of the world. So, the US was winning by getting other people to um buy into products being made in the US. So, this is a a story of the US winning not at the expense of other people cuz they were certainly >> we had the the biggest two economic miracles were not the US. They were actually They were actually Japan and Germany. They were the economic miracles that people have forgotten about that. But up until about 2000, well, in Japan's case, >> sorry, you're you're going to have to define what you mean by economic miracle because they were ultimately buying the things that we were making. >> No, they were making a No, they made the things that the US bought with the money that they made from make selling goods. It's not a fixed pie. You keep thinking everything's a fixed pie. It's not. So, the US made goods that sold abroad. Their consumers consume goods cuz they had a massive baby boom. >> Mhm. >> We bought The US bought massive amounts of Japanese goods, extraordinary amounts of Japanese goods. So much Japanese goods that the Japanese have run a surplus entire history since World War II. And they did the same to Germany who run a surplus cuz everybody buys German goods. So, here's a here's a symbiotic relationship where the US makes money as the dominant global force. They make goods that other people buy. They run a deficit. In return, they're buying massive amounts of goods from Japan and Germany and it all worked. And everybody had an economic miracle. Everybody reduced their debt load. >> Help me update my mental model. So, you're post-World War II, US is the global manufacturing hub of the world. Japan is going to come online, but they're certainly not online in the '50s. Uh I don't know the timeline as well in Germany, but I'm going to guess it's something similar. So, we begin to um flip the script and call it the '60s where this begins to reverse and Japan starts becoming the producer of goods. They become the cheap labor, etc., etc. But the period in which we start um solving our debt problem is the '50s and '60s. This is before Japan solves that problem. But hold on, because I have to get to something or I'm going to be confused forever. Uh so, the the um question for me is that if we are doing financial repression very specifically in the '50s, we are holding the yield curve down, but the only reason that it works is that yes, inflation is higher than the yield that bonds are paying, but the real economy is growing. Being a worker in a factory in Michigan, your rates are going up faster than we're inflating. And so, that's because we're making and selling, not because we're buying. We have not become that economy yet. So, what I'm trying to figure out is, okay, we have $40 trillion in debt. It's growing. It's going to keep getting bigger. We're going to run something very similar now. Warsh is saying, I think there's a lot of interpretation in what I'm about to say and you'll correct me where I go wrong, but Warsh is saying, "Hey, the the productivity miracle is going to be artificial intelligence. It's going to through an increase in productivity, it's going to make everything cheaper, but dear um human, I'm only going to let you capture some of that. I'm going to pump money into the system. I'm going to inflate the money supply and this may be where I go wrong, but this is my vision of him. I'm going to inflate the money supply to make sure that the politicians are happy. They can keep giving things away for free and I need to inflate the debt away anyway. And okay, period. So, you're saying, "No, no, no. What he's talking about is >> talking we're talking different time horizons. Yes, I completely agree. He will definitely do that cuz there's no way of doing it in the interim. You can't just turn productivity on like that. It takes time. We we'll start to see it pick up. Will the GDP growth pick up massively? Not yet. You know, maybe we peak out in this cycle at 4% which is will be good, you know, um but it's later after 2030 where it starts to matter less. So, Warsh's job is exactly that. It's just keep the ship afloat. Try not to raise rates too much. Try to give liquidity when it's necessary. Try and make sure you pay the interest on the debt and that requires printing money. Try and free up regulation on these um technology companies. Let them try and innovate and move as fast as they can. Um that that's the game. It's exactly what Greenspan did. Exactly. In the in the late 90s. He basically came in, cut rates once, twice, and did nothing. He did nothing and let productivity do the rest. So, we even had headline CPI was was high in the late 90s, but he didn't flinch. He just said, "Well, core CPI is not going up because this force is deflationary and productivity is picking up." And Walsh Trump, Bessen, and Walsh have all referred to the Greenspan period. That's the playbook. Um so, yes, you're right. He will do that, which is why stocks keep going up, why people the K-shaped economy continues, because people can't buy a house and they can't, you know, get the assets that they want and it's it's it's the same. That doesn't change, not yet. >> Okay. You You and I have talked about this before, and I think you're pretty famous for this at this point, but the the idea of the everything code, um super shorthand, liquidity drives everything. So, help me understand. I think of liquidity as, and again, these are gross roundings, but I think of liquidity as being primarily driven by debt, uh eurodollar debt being probably the most important in terms of the economic respiration of the world. Um and so, when I think about, okay, how do we meet the liquidity needs of um the increased economic activity that AI is going to bring, um I don't think of it as a fixed pie at all, but I do think there is a very critical difference between wealth and money. So, I've got wealth easy enough. Wealth can sort of come and go in an instant, doesn't matter. But when something has to be translated into money, there you can have some weird things happen. And so, I may just be confused, but I don't think velocity of of money is the same as increased liquidity. Am I crazy? Is that the same phenomenon? >> Yeah Yes and no. So, increased liquidity, liquidity is just think of it as the total supply of dollars in the system. >> Mhm. >> Simple as that, right? >> Yeah, so not influenced by how fast it turns over. >> No. >> Okay. >> That's liquidity. Doesn't matter how fast it turns over. It could be stuck. And in many cases it was stuck for a long time in the banking system and other places where they needed the liquidity or in the foreign bond holders uh the the foreign dollar it debtors in Asia and stuff like that. They suck in all of this capital cuz they need to pay the debt. They suck in the dollars, which is why we keep printing more cuz the debt, as you say. Velocity of money is how fast it turns around. So I get my $10 and I go to your shop and give you $10 and you quickly go across the road to to give it to the greengrocers and buy, you know, some fruit across the road and they take it and go and put it in their gas. Right, that's four people have used the same dollar. That's velocity. >> Yep. >> So if there's velocity of money, it's free to move around quickly and easily. So it's it's a game of musical chairs that's super fast. If there's no velocity of money, then I give you the $10 and you just put it in your piggy bank and don't do anything for 3 years. Right, that that dollar cannot be used again. So velocity is a crucial concept. Now, some people think velocity is can be inflationary. It just depends on all the other backdrops, but what you want is an economy where money's moving around super fast and it's plentiful. >> Brother, leave us with like one final. So if I represent the doom side of the equation, um if you had to condense everything that you've been talking about here cuz you've given us the breadcrumbs, but put it together in terms of how you're thinking through this moment. You're jubilant. Uh it's clear that none of the concerns have um rattled you. So if you were going to infect people with that optimism, how would you put the pieces together? >> I I think I should infect someone with the optimism because, you know, that's my opinion and I do my work and it's not it can be right, can be wrong. But this is the greatest time to be alive. We are seeing the fastest changing technology to humanity that could ever exist. This is something extraordinary and I keep saying it, we're putting electricity through sand and creating intelligence that's a million times faster than than putting it through a brain, which is carbon. It's extraordinary. And of course, that's going to attract all the capital in the world. It's the same with blockchain technology. These technologies are fundamental substrates of the new world. So, don't avoid investing in it because it might be racy. It might be something you don't understand. Maybe there's some risk in it because you will definitely miss out. And the way to deal with that is what we talked about before is how do you construct a portfolio that allows you to have your umbrella at the front door or be the person who jumps out in swimming you know, in swimming costume out the front door every day expecting it to be hot and sunny, right? It's just construct a portfolio that allows you to do that without blowing up. So, never get so scared that you don't invest or too cautious cuz you will never get out of the cave. You're [ __ ] You are always in the bottom of the cave. There's no way out. And don't on the other side with the sunny disposition, don't think I'm going to use leverage to get there faster. Tom and I talked about this. We talked about it over crypto before it blew up, during it blew up, everything. We kept explaining to people, don't do this. And then it doesn't matter and nobody can take your crypto away from you, but they do if you've got leverage. So, you don't leverage into it. So, then it's just a matter of where your own psychology is. But, just step back from all of the debates and all of the stuff and realize what is going on. We've got a single man who owns the largest private communications network on Earth, but it's in space and he sent it there by designing the rockets himself that come back down to Earth and land. He has cars that drive you. And most people haven't been in a Tesla self-driving car, put in the address, and chatted with your friend in the car and not without even having to look at the road, really. It's unbelievable. And it's This is the worst it'll ever be. If you haven't tried building something with Claude or ChatGPT, something that's in your head, could be a game, it could be a dashboard, it could be your finances, it could be your stock market portfolio, and realize that you too can now program anything in minutes by just talking to it. I don't even type anymore. I just use WhisperFlow. I talk into it and I have six Claudes open doing six different things that hurts my brain, not on the same I'm trying to fix one thing over here, write an essay over there, do some all at the same time just by talking to it with a microphone. Hey, can you do this? Hey, what happens to that? Do I need to fix this? What does this mean? Like an orchestrator. These are the times we're living in. We're living in the most extraordinary thing. So, just embrace it cuz it's not going to stop. Whatever you say, whatever anybody does, however people complain about it, it can't stop. It won't stop. So, the best thing is just to lean into it and say, "You know what? We're going along for the ride." And as you as we've talked about, Tom, that doesn't mean you have to risk all of your savings or anything else on it. You just have to realize where we're going and it's not going to stop. >> I love it. Brother, always an absolute pleasure. Where can people follow along with you? >> Um, you can either find me on X. Uh, that's easy at Real Vision or you go to my There's got a personal website, Real Vision, r a o u l p a l. Um, .com. As simple as that. Oh, by the way, Tom, the other thing that's happening is I am In the next couple of weeks, I will release out to the wild my AI in my voice, trained on all of my content. >> Let's go. >> And everything I've done and all the economics thesis and stuff, so you can just bring my AI on and have a conversation and it's in my voice. >> That's dope. And that comes out when? >> I mean, it's it's basically out now, but I just need to do a few tweaks, so it'll be realvision.ai. Um, and um, it'll be out in a couple of weeks. And it's free to free to use. You can talk to me, ask me questions, give me charts of the market. You can, you know, give me give me your doom theory and say pick it apart. It's all It's It's there and it's wild. I've even trained it on like wines that I like, travel, music stuff. It's like It's It's very cool. >> Wow, it's wild. Well, I can't wait to try it. I'll give it I'll give it some doom and some charts and see what it says. >> Perfect. Good to see you, my friend. >> Awesome, man. Same to you. Guys, if you have not already, be sure to subscribe and until next time, my friends, be legendary. Take care. Peace. >> If you like this conversation, check out this episode to learn more. >> I think America's gone too far in privatizing everything. Election campaigns are something that should be funded 100% by government money and private donations should be banned. Not even possible. Set it up in such a way that you guarantee you can't buy your