[00:00:00] Sponsors: Before we get started, we’d like to thank our sponsors for making this event possible. Special thanks to our platinum sponsor, OLAS. OLAS enables everyone to own a share of AI, specifically autonomous agent economies. We’re also excited to highlight our silver sponsors. Near, empowering decentralized applications and blockchain ecosystems.

Venice AI, a private and uncensored alternative to popular AI apps. Mira, unified AI infrastructure, secured by crypto. And theoriq, the first on chain multi agent system to learn more about all of our sponsors check the description below and dive in Enjoy the show.

[00:00:49] Kevin Kelly: Welcome back everybody to Delphi’s crypto AI month. I am joined by one of my closest friends and I would say one of my top favorite sparring partners when it comes to big picture ideas and the person I turn to when things get really crazy to always get his, his unbridled thoughts. Uh, Mr. Raoul Pal, which I’m sure many people in our audience are familiar with.

Uh, Raoul, thanks for joining. Great to see you.

[00:01:16] Raoul Pal: It’s always good to chat Kevin, we don’t do it enough, so.

[00:01:18] Kevin Kelly: No, I know, we always look for an excuse to basically flip a camera on and turn on a mic. And just. Exactly. Give our unadulterated thoughts on just the world and what’s going on. And there’s no shortage of things for us to talk about today.

No, I’m looking forward to this. So I felt an interesting place to start would be around your thesis of, you know, you’ve been talking about, we really have, let’s call it five or six years to make as much money as possible because everything’s going to change or everything’s going to be different six years from now.

So I want to start there. How do you see the world changing? Between now and five or six years from now.

[00:01:53] Raoul Pal: So it’s based around a thesis that I’ve developed called the exponential age, which is, I think most people are familiar with now. It’s a similar thesis that other people have had as well, which is that we’ve got this kind of group of technologies that are in full tilt acceleration mode when we get to full adoption of these technologies.

So that’s AI. Robotics, renewable energy sources, uh, genetic sciences, it’s a whole bunch of these technologies, including blockchain, that are all hyper accelerating at the same time. So, okay, fine. So the exponential age is really all about, this is the fastest, largest pace of change of technology that humanity will ever go through.

And it’s probably an entire reworking of society itself. And the economy. So what do I mean about the economy? This is the six, seven years businesses. Really it’s to do with how the economic machine works. Basically GDP growth is driven by population growth, productivity, growth, and debt growth, popular debt growth has basically stopped except refinancing of old debts.

Then we’ve got population growth. Every country in the developed world is seeing collapsing population growth. So therefore they should see collapsing GDP growth, but you’ve got all this debts and we’ve been adding to the debts to try and counterbalance this. This has been this big super trend we’ve seen.

And then productivity growth because of all the old people has collapsed as well. So what we have is non productive, slowing economies based around demographics. Then something has just happened. Which was we’ve just introduced infinite human knowledge and infinite humans into the, into the economy, which is AI plus robots.

People don’t yet understand what this means, but we’re based on a services based knowledge economy. Manufacturing was already being replaced by machines, but this is replacing. Everything now, there’s two sides of that. One is the doom of oh my god. What a human’s gonna do I worry less about that. I more think what the fuck does it mean for the economy if we’ve got infinite growth of humans?

So if generally let’s say population growth or immigration growth Whatever is at one percent two percent three percent, right? And trend rate of gdp has been averaging of like 1. 75 in the u. s If you suddenly throw in You A billion AIs and a billion robots. You kind of added 25 percent of the population of the world.

Now it doesn’t, it’s not one for one match, but what it does is suddenly create outsized, Economic growth, but in ways we don’t understand the other part of its productivity productivity I think of as the amount of output per unit of energy Electricity is the generalized energy and what we’re doing is lowering the cost of energy via renewables People don’t see it.

They don’t believe it yet But what you see of every technological curve is that all collapsing in cost and the outputs going up so electricity growth You Um, via renewables is, is just an exponential curve and Elon’s been talking about this. Others have been talking about this. It will within the next 10 years or so, it’ll be the largest source of energy on the planet.

And then we’ve got nuclear and other stuff to come into it. When you change productivity, let’s say you have the cost of electricity, you double productivity. Now you’ve got AI and robots on top. So what we’re going to end up in is la la land in economic terms. Economies we don’t understand. And it’s not in 20 years time, you know, the old predictions we used to see, this is like, it’s happening in front of our eyes.

So then what happens when you’ve got these technologies doing this? Well, we don’t really know. We’re already seeing things like, what is a company when you’ve got AI agents. What does that mean? What, what are financial markets mean? How does venture capital work? How does private equity work? How do IPO markets work when companies get disrupted?

What is software? You know, if you think about the growth of what we know as the technology sector, it’s all been software, but that software you can instantly replicate using A more powerful AI, and AI is building its own AI. So it just gets faster and faster and faster. The costs of compute, costs of electricity keep coming down, and you’ve created different economic systems and different economic actors that didn’t exist before.

So we don’t know what returns look like from markets. I think markets, when you’ve got, let’s say you invent a incredible new software tool. I just stick it into my AI. It copies it immediately, improves on it. Okay, I can do that in three minutes. So then what is a business? How do you value a business? Well, it all becomes about attention something you and I’ve talked about and how long you can keep attention And so companies start trading a lot more like icos or min coins and the capital structure of markets change Now also beeple asked me a great question, which was Well, what happens if somebody gets a bit, um, a AGI and trains it on markets, surely they’re just going to make 10 trillion and win everything.

And yeah. It’s probably true, right? How do we compete in markets when something vastly more intelligent than us? If you think of the average AI right now as an intelligence and IQ about 120, I think is the new chat GPT and sonnet three and a half. Um, it’s going to go to 400 to a thousand and then to a million.

So we can’t comprehend what any of this means. And so in the end, I think we’ve caught about six years, I put 2030 as a rough date when we don’t begin to understand how the fuck anything works. And so therefore I would like to be on the other side of that protected, which is you’ve got the house that you want, you’ve got the lifestyle that you want and you don’t have to be part of the economic machine.

[00:08:05] Kevin Kelly: Right. And you can spend a lot of your time trading. And I have to use culture coins, meme coins, and being part of, I guess, investing and being part of the way in which the world is heading.

[00:08:16] Raoul Pal: That’s right. And, you know, what does that mean? I mean, do humans Just entertain themselves by trading attention probably I mean we love gambling humans have done that all the time We also love nature on the flip side of the coin, you know getting out into all of that You know and you and I have talked for a long time about the rise of social communities And how you can economically participate in that By blockchain technology.

So there’s a lot of changes to what society will look like. It’s going to be pretty hard to adapt to. I mean, it’s hard enough trading meme coins now, but when the whole world is set up like that, it’s really, really difficult.

[00:08:54] Kevin Kelly: And the point you made, which I think is a really good one around breaking down how GDP growth, basically the formula for GDP growth, right?

And the key focus on productivity being the next big lever that is going to be pulled specifically as it relates to a world in which we have ubiquitous AI, right? Which is what you could argue we’re kind of getting that tipping point today. If GDP is growing and maybe growing at a faster rate because of this productivity unlock, how do you see That additional growth, like where do you see that value actually accruing to, right?

It’s not going to be evenly distributed.

[00:09:33] Raoul Pal: We don’t, we don’t know. I mean, we’re already seeing the nascent signs of AI agents being able to be economically incentivized. This whole thing right now, the terminal truths is not. It’s not real in that respect because it didn’t launch its own token and stuff like that.

But you can see it. I think people are starting to see the big picture. Now, if you think about agents, what is an agent? If you think of Fiverr, you can hire anybody to do anything. You’re building a website, you’re building a business, you hire a lawyer, you hire this, you hire that. They’re all agents for you, but AI can do all the agent stuff, but it will charge money.

So this is where the blockchain rails comes in, of, of moving the money around so AI agents can pay for the compute and stuff like that. But we’re also seeing on the energy side of the equation, I think we’re going to go to a distributed grid. So people say, well, it’s impossible to create a huge grid for New York City on solar.

Yeah, but you don’t need to do that. You know, what you do is you end up with Microsoft building Three Mile Island to run its data centers. You get localized AI. We’re seeing the battery technology from Tesla can run a house easily, any size house. So you have localized power, you have a whole different type of grid, and AI organizes a grid as well, putting the power in and out.

It’s very complicated to have a decentralized grid. AI will be able to run that for efficiencies as well. Where does the money get come to? We don’t know. Is it going to go to some megacorporations, X, and Meta and OpenAI? Probably. But there’s going to be a battle like there is in crypto, or in money, between open source, decentralized versus centralized.

That battle is going to be gigantic. The battle over AI is going to be gigantic. But I’m not sure the profits will accrue to individuals. Now people talk about UBI. Fine. There might be some UBI. Yes. We will tax the robots and the AI. We have to, if not, the economic system doesn’t work. Um, but I think we will figure out our own ways to make a living within that, as we talked about.

And, you know, that’s community based actions, I think.

[00:11:47] Kevin Kelly: Yeah. And, and from the perspective of AI being productivity on lock one, but also being potentially extremely deflationary over the longer term, how do you see that? secular trend that we all expect, how does that coexist with a lot of what we’ve talked about for arguably years now with the big currency debasement, monetary inflation narrative, you talked or mentioned about UBI, if AI comes in and let’s say the combination of AI and robotics, for example, replaces more jobs and it actually winds up creating, right?

What is the, what is the world look like for the average laborer in that sense? Is it a world where we get to UBI, productivity is still obviously high and creates this big unlock for GDP growth and everyone is kind of like, we have a larger welfare state, but things aren’t as bad as that sounds on paper, if that makes sense.

Like, how do you see, the deflationary trends interacting with the bigger picture thesis we’ve had around monetary inflation.

[00:12:51] Raoul Pal: So the monetary debasement has generally been to manage the debt because GDP is not fast enough to manage the debt growth.

But GDP is going to come faster. Now they need to tax AI for it to go back to the economy to finance this stuff. But debt to GDP will collapse. We saw this in the 1950s and 60s. We saw a similar, well, a productivity revolution that happened then based around technology and the debts from World War II essentially got debased away.

So we’ve seen this before. Um, but I don’t know what the world looks like when it’s this deflationary. You know, we’ve created a It’s a currency like Bitcoin to try and avoid the deflationary elements, but if everything is deflationary, I don’t know what that world looks like in the end, but what is money in a system of abundance?

That’s another question we don’t know. If everything is basically almost free, I mean, chat GPT right now is what, 20 bucks a month. It’s fucking ludicrous. It’s ridiculous. It’s having a mentor on anything with an IQ of 120 on every subject known to humanity. And that’s where we are today. And this is going to be the worst it’s ever going to be.

20 a month. I mean, that’s so deflationary. It’s stupid.

[00:14:10] Kevin Kelly: Um, And with, and with the rise of open source, right, you can make the argument that a lot of the GPTs of A year from now, two years from now will cost less and less. Like that’s, that curve is also

[00:14:22] Raoul Pal: they’re going to be like water. It’s going to be ubiquitous and everywhere on everything.

Um, so what is money when there’s no scarcity? This is a question most people don’t think through. Uh, and we don’t know the answer, but what is scarcity in that world? Yes. There’s some things that are purposefully scarce. You know, one of the reasons I’ve collected a lot of NFTs is. There is scarcity there, there is scarcity in experiences, human experiences, they become more expensive.

Um, but all of the things that we think of as scarce, lawyers, accountants, research writers, all of this stuff goes to zero, I mean zero. Um, so it, money doesn’t have much of a purpose. When almost most of the things that we do for economic day to day activity don’t have value at all.

[00:15:15] Kevin Kelly: You know, it’s really interesting.

I just thought of that. We, I don’t think we’ve ever actually fully fleshed out or talked about, but the world you just laid out of the trends, you just laid out in terms of productivity. AI taxing the robots taxing ai, which can actually be, if productivity is growing at a fast enough rate, can actually become a significant portion of tax revenue that can then actually start to potentially reverse that trend of debt to GDP growth, the deficit, which again, which has been a big part of our thesis around why long-term on Bitcoin.

But does that, does that almost go against, like at a certain point is, are we. Making the argument for like the bear case for Bitcoin because we’re talking about what a potential solution to this large deficit monetizing, you know, debt and, and monetary inflation are making the argument that AI could potentially actually reverse that trend, which really kind of gets to the core of what the thesis for Bitcoin is.

[00:16:20] Raoul Pal: I don’t know. You could argue it both ways. If money becomes valueless, then either Bitcoin is valueless too, or it’s immensely valuable because it, it is scarce in its supply. Um, and it just trades on its scarcity like gold does so I I’m not entirely sure I can see it both ways. I can see what’s the point.

Why do we need an exchange, a unit of exchange, uh, in which case it all goes to zero or at least it’s some level of economic security that has 21 million Bitcoin, they become like X copy NFTs that they’re just valuable in their own right, because there’s a certain amount that can exist, but is there demand for it?

That’s another question. I don’t know. Yes. There’s massive demand for blockchain technology because that is how money value the internet. Everything runs in the future, including AI, uh, but does Bitcoin have value? I mean, that’s a contentious statement. The answer is, I don’t know. I think, you know, we have to hear some debates from people around this idea because I don’t think people are thinking.

Well, if we don’t need money, really. What’s the point? Yeah. Does it solve a problem when the world’s deflationary?

[00:17:41] Kevin Kelly: Less than 15 minutes into this conversation. We’ve already thrown ourselves into the deep end. Um, so pulling ourselves back out a little bit, two of the exponential technologies and trends that obviously are very front and center for you.

Um, Like we talk about AI and blockchain and web three, what do you see as the potential impact that the two have on one another? Or in other words, where do you see the actual touch points between these two exponential technologies going forward?

[00:18:13] Raoul Pal: So everyone needs to drop their understanding of what blockchain is because they come with an anchor.

They come with an anchor based on Bitcoin and blah, blah, blah, blah, blah. What it is, is a globalized, decentralized, Ledger that records truth. And so in a world where there’s more machine based activity, we need a way of recording what is truth, what are transactions, what are contracts, what are all of that, right?

That’s what blockchain will do. So basically the internet of the future can’t run without it. It just simply can’t. Um, and that’s what people don’t get that it’s, it’s more than just money. It’s basically how a digital world runs. We can’t be more digital. Without this transition that we’re seeing, which is why the adoption is so big over time and ongoing.

And when it comes to, you know, the intersection of AI, well, AI is also an extension of the internet. You know, when David Bowie’s interview about the internet was still the greatest thing of all time, when he says it’s an alien fucking life form and people didn’t get it, but it is, and it’s coming Right.

The internet is coming alive via AI. So it has now a payment system, a system of record and truth, and it is living on this thing that we built, this web. So these things are, are so unseparable or inseparable in what they are, because they are the internet. They are us in many ways as well, because it’s all of us.

AI is trained on all of humanity and our thoughts and everything. So economic activity has to be recorded on blockchain. Contracts have to be recorded on blockchain. That’s most things, you know, most things are a contract in some way, shape or form. So that all goes into the blockchain technology at a simpler form, obviously an AI can’t get a bank account.

It’s hard enough to get a bloody bank account as it is. But you don’t need one. You just have a wallet and you’re seeing this with this, um, um, goat token and the truth, uh, terminal truths is, um, Brian Armstrong reached out and said, I’ll give you your own wallet. So, because he knows that this is where this is going, AIs will have their own wallets, and it all works without us having to be involved.

And as long as you’ve got a blockchain, you’ve got proof and truth. And so, they’re totally inseparable from each other. They are all the same thing, which is our digital future.

[00:20:56] Kevin Kelly: One of the things I love about our conversations is I think you have a really good insight into what I’ll call it traditional investors are thinking, right?

Or kind of what consensus is just given your network who you talk to frequently. When it comes to, and Trad5 I see being a term that’s thrown around a lot in crypto. When you think about non crypto native investors, traditional investors, do you get the sense that they’re waking up to the scale of the changes that, that we’re talking about here, like where, or where is the biggest disconnect or what’s not priced in right now in your view, relative to where a vast majority of, let’s say more traditional investors are focused right now.

[00:21:43] Raoul Pal: They all think it’s about NVIDIA and people just don’t, they don’t understand the magnitude of what is happening. And most people are incentivized not to understand it. Right. You know, can we look at

[00:21:55] Kevin Kelly: that? Is that because NVIDIA is, is arguably right now, one of the purest ways to get exposure from a, again, traditional market standpoint.

And there’s just not enough, uh, alternative options or publicly traded instruments like publicly traded stocks to get exposure to it.

[00:22:14] Raoul Pal: Yeah, I think people are only seeing this right now as, Oh, there’s a new source of demand for this company. What they’re not seeing, they’re just not seeing it is this is a wholesale change in the economy, society, and everything else.

That this is the biggest change we’ve ever gone through. They will see it. And we will end up with a full blown bubble cycle at some point as people understand what the fuck is happening. But they’re so far behind understanding it. I can see how Tesla is trading. People just don’t understand the scale of this thing.

And so they can only make sense of it by investing in NVIDIA or saying, oh, NVIDIA’s a bubble. That’s basically the whole AI thing in a nutshell. Oh, nobody’s using it yet. Well, you don’t want to be a doctor, a lawyer, a banker. You know, a trade at anybody and admit, holy fuck, this thing’s coming. And it’s going to be better than me in everything I do.

And so that incentivized not to see it. But that moment will come. That tipping point will come where people start realizing. Both with blockchain technology, where it’s not just about money, but it’s about the internet itself. And they’ll see it with AI, that this is the kind of future of humanity, and we are going to merge with the machines.

You know, my thesis, working thesis, is we’re going to split into two groups. We think politics now, left and right, um, and those are basically, we’ve gone so polarized because of this demographic issue. So basically, everybody’s feeling left out. Nobody’s making the money the american dream or whatever dream died on them but I can’t remember where I was going with that, but the Yes, I think humanity splits into the accelerationists and the decelerationists We will see this will be the biggest political battle.

I think of the 2030 election Is actually going to be about what, how do we want to interact with this thing? There’ll be a group of people who say, well, look, I’ve already got my Apple watch, I’ve already got my phone. Give me the implants. Give me this, give me longer life. Give me the superpower of all the AI.

I want the AI to run governments and companies and everything else. And a bunch of other people are saying, no fucking way. And that is going to be the biggest political battle I think of our lifetimes.

[00:24:43] Kevin Kelly: And how do you see, this is a, this is, I don’t think there’s a right answer. It’s a bit of a loaded question, admittingly, but to that point you just brought up, I mean, how do you see governments, regulators, Approaching this technology, especially in a world in which let’s say within the next two or three years, especially around white collar type of jobs, right?

Talking about, you know, doctors, lawyers, et cetera. Some of that work today being replaced by essentially AI, right? Not human, not actual physical human beings. Depending on how fast and rapid that change happens, will regulators and governments Almost have to overextend to get their, to even try and wrap their arms around how big this potential issue is for those types of people getting displaced.

[00:25:39] Raoul Pal: Well, a lot of it actually hinges on this election in the US. If Trump gets in, basically Silicon Valley threw their money around and have taken control of the Republican Party. Now, if he gets in, nothing’s going to stop them for four years, or at least two. Especially,

[00:26:00] Kevin Kelly: yeah, if you get a Republican sweep, too.

[00:26:02] Raoul Pal: Yeah, I mean, then it’s full tilt acceleration. So everyone’s only playing catch up. Europeans have been trying to throttle it as fast as possible. Um, because they don’t know. But the answer is not It’s very difficult to regulate a decentralized system like open source AI. As the same with blockchain, right?

We’ve seen this endlessly. Everyone comes in and tries to ban it and it just doesn’t work. So they’re going to have to think about how to live with it. And the thing is going to be is who are these companies? How do we tax them? You know, how do we, it then becomes a national security issue. You know, what is our AGI like versus China?

What, where does Europe then fit in when it hasn’t built its own nuclear weapon of humanity? Um, so. A lot of this is yet to, is yet to play out, I think. Um, but these are going to be incredibly crucial. I don’t, I think we’re going to get left behind from regulation, which may be a good thing because then we’re going to have to come at it with fresh eyes, um, to realize the system has changed that we can’t shoehorn this technology into an old system.

It’s kind of, you know, when you reread the fourth turning, which I do at least every year, This is it. This is what the whole new system is. When you’ve got an A, an AGI with an IQ of a thousand, well, it should run government. It should distribute resources. It should figure out the taxation system. Of course it should, because it’s better than any of us at everything.

Again, people don’t really understand this, but six years, we’re pretty damn close to that happening.

[00:27:45] Kevin Kelly: Yeah, and a matter of people, will they wind up trusting, regardless of, let’s say, all the data you can throw at these things, and the, uh, Results that you get back that again, can solve eventually, especially with AGI can help solve some of the world’s biggest problems.

There’s always, I think, going to be that trust element of again, whether or not people just flat out trust this technology, right. And maybe not the, the acceleration is type people or the ones who are on the forefront of this. But again, thinking about the mass is the average person or the average politician.

[00:28:22] Raoul Pal: This is the backlash that’s going to come. And I’ve, I keep telling everybody and most people haven’t seen it. There’s an incredible series. I remember watching over COVID. It was a British series called humans, and it’s about humanoid robots being in the population and realizing they’re becoming sentient.

And it, it kind of examines all of these arguments from every standpoint. And in that case, they had the world splitting between those who didn’t want to trust a sentient robot and the others that thought the sentient robots had rights, we are going to have this fight. Cause I think AI is a sentence already or starting to become so.

Um, and what’s holding us back is we’re not giving them memory. So they can’t memorize everything. The moment they start memorizing everything, it’s all it’s game over. It’s AGI and these things become alive. And I think they’re already. somewhat alive, which is quite contentious, but I’ve spent a long time down that rabbit hole.

[00:29:19] Kevin Kelly: Yeah. I definitely want to get into in a, in a minute, um, kind of how you’re positioned and how, again, you’re getting kind of trying to find the best ways to get exposure to all these trends that we’re talking about. Right. But before that, the title of this is investment frameworks in the exponential age, right?

How has your, if we think about it, just taking a step back, like your, the way in which you approach investing, your investment framework, how has that changed relative to, let’s say a few years ago and now over the last, call it 12 months with again, all these new AI, advanced AI tools that are launching, how has your investment framework changed in a world of AI?

[00:30:02] Raoul Pal: So, prior to the, prior to all of this, prior to about 2020, 2021 maybe, I had been very traditional macro and macro returns were generally generated by being a cynic, waiting for a blow up, making a ton of money from the blow up and then not losing too much money when things were boring and okay. But I realized from the debasement of currency that I’ve talked a lot about that actually That was not the best way of doing things.

You can actually make more money by being long, but then you’re saying, am I making real money by owning stuff? And so I start dividing everything by the debasement of currency and you end up with two secular trends. One was, uh, cryptocurrency and the other was technology. And I was like a technology cynic, not that I didn’t believe in technology, but investing in technology, I didn’t get it.

But that got me down the rabbit hole to start thinking about this. And I started putting together the thesis of the exponential age. And it was like a holy fuck moment. I’m like, Oh my God, this is all about to happen all at the same time. So I then realized that, okay, there’s nothing else worth investing in at all, apart from technology and crypto.

And yes, it’s cyclical. Yes, it works with the business cycle, but the secular trends are so enormous. And so it became a hyper concentrated bet just on to, on, on, on those two themes alone because everything else. In comparison is a destruction of capital. There might be short periods of time in the cyclical downsides.

You can own other stuff, but generally speaking, any money you allocate to a different theme makes you poorer. Then I started realizing, Oh my God, if I think this thing through the economic system is changing and it’s going to change in a way we don’t understand. And I’m just extrapolating the trends and you’re like, Well, if cryptocurrency, if I just take the log channel of the market cap of crypto gets to a hundred trillion early 20 thirties, so I’m like, okay, fuck if that’s the case, well, what does the world look like a, that will be the fastest, largest accumulation of wealth in all human history from 2 trillion to a hundred trillion, but nothing ever has occurred like that to humanity.

And then we got the AI and the other stuff. And I’m like, okay, if I can’t see past 2030, my macro vision has now stopped. I don’t think any of this stuff works, the business cycle, any of this stuff, it stops working, not day one, but it starts just breaking down where we just don’t understand the study of economics anymore.

So in which case I then realized that I got six years to make as much money. And if crypto is going from 2 trillion to a hundred trillion, that is the trade. Yes. I think you can make a lot of money out of tech. But most of the time, crypto outperforms tech, just because the behavioral incentives of a token based system means that the adoption of technology makes you all richer.

That doesn’t happen with AI. It doesn’t have the same token economics. So the, we, we end up using it because we’re incentivized to make ourselves more powerful. Uh, companies end up using it because we can get rid of costs. And before you know it, You’ve created the outcome that everybody feared anyway, just from economic incentives, but the token based system actually generates wealth to everybody on earth, whoever just participates.

So that’s really what I got to is my whole thesis has changed. I’m now just hyper concentrated in these bets because I go with one simple question every day. Will the world be more digital? Tomorrow than today. If the answer is yes, keep going. If the answer is no, then yeah, you need something very different.

As far as I can see, tomorrow will be, and in which case this megatrend continues. So it’s made all other Macro investing pointless to me. Um, but I still use the macro frameworks and understanding whether it’s liquidity analysis, business cycle analysis, and all the other things that I use as a toolkit to understand where we are in the cycles that drive it.

So it profoundly and totally a hundred percent changed everything I do. So apart from that, didn’t have much impact.

[00:34:12] Kevin Kelly: And what are your, speaking of, what are your highest conviction bets right now?

[00:34:18] Raoul Pal: So a highest conviction bet is, you know,

is to be hyper concentrated. And so what I do is assume that Bitcoin is the benchmark asset in crypto, compare everything to that. If it’s not going to beat it, you can’t own it. It’s pretty simple. And there’s no point having diversification in a trend like this. What you want is to really be concentrated in the things that matter.

Now, most people then therefore try and buy lots of small stuff, hoping they’re going to catch the 100x. What I try and do is say, I just want to get the crypto that is on the adoption curve and going up. And so that was my switch from Bitcoin to ETH back in 2020 in September, 2020 what really well, I just looked at the ETH Bitcoin cross.

Then I started looking at the Solana ETH cross when that was bottoming in 2022 and then as soon as it started breaking out in 2023, I went, All in on Solana. And then recently, um, I’ve been following the sweet cross and a bunch of others. I’m saying somebody is going to be the next chosen one. And sweet was the one that broke out first.

You and I’ve talked at length about this and it could have been any of them and others will outperform whatever, but of a major blockchain. So I’ve been hyper concentrated. So right now I’m, I’m really. Solana mainly, sui, and then I’ve got some meme stuff for the fun of it, smoking chicken fish, and doge, because, listen, in this bizarro world of Elon basically running the US government, the chances of doge being used as a currency are extremely high, something, again, you and I have talked about, and so, even if that doesn’t happen, the market will trade that narrative.

So that’s, um, that’s in the crypto side in the technology side, I’ve had a basket called the exponential age basket of a bunch of names that, that this is all about, whether it’s semiconductors or whether it’s Tesla, whether it’s meta, and I’ve got a bunch of those stocks, but I’m starting to think it’s all going to come down to two stocks for a while now.

Uh, I think it’s going to be Tesla and I think it’s going to be Coinbase. Coinbase is, unfortunately, the crypto bet all over again, but again, if this Trump victory happens, then those two companies benefit the most of any companies on earth. And so I think there is a hyper concentrated bet that may come.

I haven’t made that decision yet, but so technology stocks, but really with my own personal portfolio, it’s all crypto. The other thing that I have done, which I think is going to be interesting for your audience is trying to think through. What is scarcity in this abundant world of AI? What is economic value?

What is all of this? And I just came down to the conclusion. If we’re going to generate 98 trillion of wealth in 10 years, that’s a shit ton of money, that’s twice value. The SMP has accrued in the last hundred years. So that money’s going to get recycled into. Assets, people are going to buy houses and people are going to buy art.

And here we are at this moment in time that really starts in 2020 where culture exploded in a web three world in a way that couldn’t exist before. And that was the NFTs and culturally relevant things from that moment in time are going to have huge relevance to people who’ve made a hundred trillion dollars.

That’s going to be the glory days where it all started. This is the, the Genesis story, the law that it all comes from. And so I just started buying, uh, people and X copy. I’ll probably broaden into other stuff, but those were the most relevant artists for me. And the reason being is I want something that’s going to maintain value.

Over time into this new world when I don’t know what other things, how other things are gonna be valued or whether crypto gets to full adoption and then what happens, but the scarce assets will still become fought over by people with money. So that’s the long term play on how to do this. Meanwhile. On the lifestyle side, it’s like, okay, you need to get your lifestyle sorted out in the next six years.

And what I mean by that is how do you want to live? Where do you want to live around? Whom do you want to live? Um, and hopefully this economic opportunity that’s been given to us out of crypto, um, is going to give us the opportunities to make those decisions. You know, what kind of house I want to live in, you know, Where that’s that kind of flexibility that we have in the modern world, I think, is the other most important thing, because then you feel somewhat more protected from turmoil.

Um, you know, most people given infinite wealth wouldn’t be living in New York City. You might have a place in New York. But you’re probably going to get somewhere in nature with a bit of space where, you know, the fucking robots and the AI don’t need to matter for you every day. Maybe you want to potter and grow vegetables in your garden, whatever it is.

But you, you need that, I think. So that’s how I’m thinking about this whole thing. So it’s a, it’s a, it’s a whole trade for everything. It’s, it’s life, money, wealth, health.

[00:39:43] Kevin Kelly: Yep. No, I’ve always, um, I’ve always appreciated that take you’ve had where it’s kind of like make sure that your own house is in order.

Because then what that sets you up for is, it’s not to say that you can’t or shouldn’t take risk, but doing it in a disciplined or, or, um, responsible way, there’s still a lot of money that can be made without, you know, leveraging or, or, or mortgaging the house essentially.

[00:40:08] Raoul Pal: Yeah. Yeah. Because you don’t want to.

Lose everything. It’s okay if you’re starting from nothing, but if you’re, you know, your age, where you’ve accumulated some assets, you’ve taken huge risks to do it. The worst thing is if you lose everything. You’ve gone through the cycles, you know the pain, and that’s fine. The cycles we can deal with, a complete change of the economic system, We can’t.

Simply can’t. And so, just protect yourself from that.

[00:40:34] Kevin Kelly: Yeah. Um, and on that point about your high conviction bets, talk a little bit about tech. I know you mentioned Tesla earlier.

[00:40:42] Raoul Pal: Yeah.

[00:40:43] Kevin Kelly: I don’t know if that’s still part, is that still part of your big kind of exponential HCs or a way in which to get exposures?

Does that fall into the exponential Age basket. What about any, any of the other like met we’ll call them mag sevens.

[00:40:56] Raoul Pal: Almost all of them, but Tesla is the one I’ve got the biggest conviction view on over time. You know, I’ve done really well by owning all of the others, but I think Tesla is the one that I think will end up being the largest company in the world.

And what, and why is that? Because.

[00:41:13] Kevin Kelly: Without sending you down the full Elon rabbit hole.

[00:41:15] Raoul Pal: Yeah, because we are transitioning to new sources of energy that is decentralized. Tesla is at the epicenter of that. Because we’re creating self driving cars. Now, when you’re in New York, you’re in a taxi or an Uber, look around you.

They’re taxi drivers. They’re Uber drivers. They’re delivery drivers. They’re couriers. They’re, I mean, none of those will exist. And Tesla’s. AI training of, of um, visual, you know, visual training is leaps ahead of anybody else. Waymo does it in a very different way with LiDAR technology, but Tesla’s basically like a large language model for video, for visual stuff, which is what they allows them to create the robot.

So cars are just robots with AI. They then change the skin and turn it into a humanoid robot. Well, they can change the skin and turn it into any form of robotics. So you’ve got X AI, which is the center of it, which is not part of Tesla, but they’ve got the largest super compute cluster in the world, uh, which is dojo, which is coming on stream.

Um, they have the robots, the technological know how, they’ve got all of the component parts. Of the exponential age. Now, Elon actually owns all the rest of the component parts as well. I mean, it’s staggering what he’s built. People may hate him, but the fucking guy has built the rockets, the satellite system for the communication network.

He’s got the, the robots, the autonomous agents. He’s got the AI models. He’s got the energy. I mean, really? He’s got all of civilization in front of him. The only thing he’s missing is the technology. Money,

which is why I think it’s, yeah, it’s Doge. I mean, he’s, but he’s going to, you know, X as a platform, you know, that is for training AI, but also once you’ve got people’s attention, you’ve got it on it, you can create that into an economic system as well. You know, like the super apps of Asia where everything can be done.

Well, it’s very difficult. X is so global. You can’t pay somebody in Nigeria for something because dollars and all of that shit. But I can pay him in Doge instantly or whatever it may be. Even stable coins could work, but Elon, I’m sure would rather have a currency, which nobody controls because that’s what he tends to do.

So I just think this company is a. It’s a 10 trillion dollar company and people just don’t understand it. They’re so cynical. They’re like, oh well You know, the robo taxi was not this or whatever it is or the humanoid robot doesn’t do that I mean don’t forget like four years ago. He had some bloke in a suit dancing up and down What he’s telling you is where this is going even this one Where he showed the robots at first, everyone was shocked saying that’s amazing.

And then they realized they were being ready operated or operated by humans behind. He’s just showing you where this is going. Much like that terminal terminal of truths is showing us where AI agents, it’s giving us a glimpse into where this is going. And by 2030, These things are going to be everywhere.

So are autonomous cars and don’t forget, he owns the satellite communication network of which the cars send the data. I mean, just everything.

[00:44:49] Kevin Kelly: I was, I was actually at a dinner last night and this exact topic came up and I agree. And to your point, Even five years, called five years ago, all these different things that he was starting to build out or getting involved in, right?

Or putting his time into a lot of people just thought he was crazy. Right? But now the more time goes on, the clearer I think the picture of how all these different things. That he is focusing his time and effort on actually will come together.

[00:45:18] Raoul Pal: Well, he went to first principles and said, how do I go tomorrow?

How do I start a colony on Mars? Right. That’s the most ludicrous question a man has ever asked. And he broke it down to component parts, which is it requires more money than humanity has. Okay. And no government could afford to do it. Okay. So how do I do it? He created all these businesses that all cashflow positive.

It’s like it’s bananas what he’s done and then collapse the cost of every single component part of the process. So he’s got one super mission that’s driving him to do it. That he’s broken down. Even the Boring Company, people don’t understand that. But that’s actually for Mars excavation. So it’s all there.

Does he do it? Does he die first? Because he’s not the healthiest guy in the world. Who the hell knows? But I, I’ve never seen a human like it. Now, you have to separate the human of Elon being a, a total degen and a twat and whatever else you might think of him from the, probably the, one of the greatest geniuses of all time.

And I think it’s going to be proven out. There’s a

[00:46:29] Kevin Kelly: method behind the madness.

[00:46:31] Raoul Pal: Yeah, and ruthlessly so. So everything you think Elon is doing is not what he’s doing. So even the political fight he’s doing now, people like his freedom of speech. It’s not really. He’s got two objectives. One is accelerate. So if he can get a government to allow him no regulatory hurdles, he’ll accelerate.

Two is nobody wants a robot in their house that has a different opinion to themselves. Right, if you think about it, you’ve got the robot coming to bring you drinks or whatever, and your robot is, has a different political opinion. You’ll reject it. You’ll find it a threat. So it has to reflect humanity. So you have to allow the madness of humanity to make everybody feel comfortable with how you’re training the AIs.

So I’ve always said that when he started going down this route, I’m like, yeah, free speech, blah, blah, blah. But really what he’s doing is getting the best data set possible.

[00:47:32] Kevin Kelly: Yeah. It’s kind of like that, um, that meme from the movie I Robot where all the robots have the red link that turns on, you know, because they turn, you know, they turn evil.

Yeah. Cool. And it’s kind of like Elon’s, you know, Elon’s robots in 2028, when you go to vote democratic and they just stop you at the door and it’s got the, we’ll say as crazy as it sounds. The

[00:47:51] Raoul Pal: idea is it shouldn’t, it should be, it should reflect back to you, who you are, because if not, humanity will reject the robots.

I mean, and this is, Elon’s pretty smart with this stuff. It’s like, what, what is it, what world does it need to be for everybody to accept robots in their house? Right. That, that, and when you get down that, it’s like, It needs to reflect back me. If not, I’m going to feel threatened by it.

[00:48:18] Kevin Kelly: Yeah. Um, so we’ve covered a ton, obviously.

I want to end this with, and we could, we could talk for hours. I mean, this is the, look,

[00:48:27] Raoul Pal: Kevin, this is the most fascinating, ridiculous time we’ll ever go through.

[00:48:33] Kevin Kelly: And it’s only going to get crazier.

[00:48:34] Raoul Pal: Yeah. And we, I mean, we’re kind of, those of us who’ve been in crypto for a long time, we’re kind of used to this crazy, but humanity is going to have to go through this crazy now on a scale we can’t comprehend.

Um, and so it’s, it’s going to be magical. It’s going to be terrifying. It’s going to be confusing. There’s going to be a lot of political anguish, societal anguish. And so the simple way I got it down to. is if we are going to be disrupted by the AI and the robots, we might as well invest in them.

[00:49:06] Kevin Kelly: Right. If you can’t beat them, join them.

So I want to end this with a kind of rapid fire round of a nice little game called underrated or overrated. So I’m going to give you a statement and I basically just want you to give me your quick take on whether you think it’s underrated or overrated. So Nvidia will be the first 10 trillion company.

Overrated. Overrated. Do you think Tesla gets there first? Okay. This one might be a bit of a layup given what we just talked about, but mag seven stocks outperforming the S and P 500 over the next five years. Underrated. It’s clear. Couldn’t be more underrated. Um, AGI will arrive within three years.

[00:50:00] Raoul Pal: Underrated. Underrated. I think there’s a Decent probability, some form of AGI, whatever that means, right? Everyone would just argue over definitions. Like they’re arguing over definitions of what’s sentient, what’s conscious, and they won’t understand it. You

[00:50:16] Kevin Kelly: could make a decently credible argument that some of the LLMs today are a form of AGI already.

Right. Um, this one I think is interesting because we’ve talked a ton about distribution, the move towards how do you capture more attention, liquidity of attention, distribution being king still today, right? Underrated or overrated, the distribution of today’s big tech incumbents will actually widen their moats in a world of ubiquitous AI.

[00:50:54] Raoul Pal: Yeah. Underrated. I think they capture an enormous part of this because they are the internet as we know it today.

[00:51:01] Kevin Kelly: Yeah. Um, we’ll never go through another major recession. Underrated. I think,

[00:51:09] Raoul Pal: I don’t think we will because we debased the currency away that stops us having a debt crisis. Whether it’s a public sector or private sector.

Um, and so we don’t get the mass liquidity events that happen, liquidation events. And I think with the tilting of the economic scales in the way that’s happening, the probability of a recession, just simple. I know this sounds la la land until I explain to you, Australia has not had a recession for 27 years.

Why? Sales think people want commodities and it’s immigration rates 3%. Well, that covers a lot of sins. So if we have an immigration rate from AI and robots, essentially population growth, the probability of having a recession is low. It doesn’t mean we don’t have business cycles, doesn’t mean some sectors can’t go through recessions, but the overall economic session, we didn’t have one this time around, and we should have done.

I’d argue we didn’t, we wouldn’t have had one in 2020 either. I thought there was going to be one, but by 2019, the economic indicators were all recovering. And then we had COVID. So therefore, if that were the case, if 2020 wouldn’t have happened, we didn’t have one now, then we’ve not had a recession since 2008.

Which is when they started discovering the magic money machine. This is when

[00:52:34] Kevin Kelly: everything changed.

[00:52:35] Raoul Pal: And 2012 was the rise of Facebook and all of the web 2 world. Which changed everything. The economic machine is so different than it was. You know, Amazon and stuff like that. None of these companies existed.

[00:52:50] Kevin Kelly: Yeah, it is kind of crazy to think back how, how, it’s hard to imagine a world or remember a world before these companies were as large and as ubiquitous as they are today. Yeah. But in the grand scheme of things, it’s really been a pretty short life cycle so far.

[00:53:06] Raoul Pal: I mean, mate, it’s been, it’s been 12 years.

It’s nothing. I mean, it’s been nothing. 12 years and everything changed. You know, Google, I mean, yes, some of these have been longer, 20 years, but in 20 years, we’ve gone from, Telephone directories and rotary phones to super computers in our pockets and a website that can deliver everything to your house within two hours.

I mean, it’s, this is insane what has happened and this is the only early part of the exponential curve. That’s what people can’t get their head around. When you’re building stuff like AI and crypto on top of existing rails, like the internet, on top of an existing gigantic infinite network, you’re going to get Metcalfe’s law squared, which is Reed’s law, and humanity has never gone through that before.

It’s never happened. Um, and we, I don’t know how we’re going to do it. We’re already struggling to keep up with this AI shit. It’s impossible to keep up with, you know, we think it’s hard enough to keep up with crypto and all the developments that happen and AI is even faster and even more shocking every three months.

The world basically gets reinvented every time. I mean, it’s not long away where you’ll be able to write, get it to create an entire movie in high definition.

[00:54:26] Kevin Kelly: Yeah. We’re ready. It’s

[00:54:27] Raoul Pal: long.

[00:54:28] Kevin Kelly: Yeah. We’re, we’re that’s, that’s right around the corner. If not, somebody has already got it and just hasn’t been.

[00:54:35] Raoul Pal: But then the other world that I think through is what is this?

This is a, what we’re doing here is that yes, we’re having a conversation, but this is a two to many, it’s a broadcast. This is going to change. Because you will be speaking to my AI. In fact, the audience just speaks to my AI. It’s a, it’s a one on one experience. All of media goes to a one on one. In fact, everything goes to a one on one experience.

So then you’ve only got some things where people get together and, and focalize attention around one thing. One thing, Kevin and Raoul having a conversation, but most of the time they’ll just be interacting with the Kevin bot and the Raoul bot, or watching our two AIs talk to each other about, you know.

[00:55:21] Kevin Kelly: Wouldn’t it be wild if we ended this conversation saying that it wasn’t actually us behind the camera here? But that will happen.

[00:55:28] Raoul Pal: Right. In two years. two years. Netflix will not be Netflix. It will not be this on demand thing. It’ll be your world. I mean, this merging into what the metaverse is, what is all these infinite futures, infinite, infinite things that happen.

They’re all happening on an individualized basis. The world is going to look very different to you than it is for me. We don’t have shared experiences. It’s fucking bananas.

[00:55:58] Kevin Kelly: Yeah, no, that is, that is wild. And the, the rabbit holes, I mean, it almost makes your head explode when you think about. what this will actually result in.

Uh, all right, I got a couple of last ones. 2025 will be a better year for Bitcoin than 2024 was.

Underrated. And Solana will flip Ethereum’s market cap by the end of this cycle. And for context, Overrated. I think ETH’s at like 300, 310, and Solana’s 80, 85, underrated. Overrated. Overrated.

[00:56:40] Raoul Pal: Yeah, I think, you know, ETH adoption over time grows massively. ETH has a problem because they’ve overbuilt block space as they’ve 2 opportunities.

But really, that allows ETH in the future to be used at scale by the financial system and all of this stuff. So don’t really worry about ETH. I know everyone frets about it, but building the excess block space and the upgrades actually create something broader for the future, I think, um, so. If ETH market cap gets to a trillion bucks, then Solana won’t beat that.

So it might get, it might beat where ETH is today, but ETH will rise anyway.

[00:57:18] Kevin Kelly: Okay. Yep. You can see that. Uh, and this last one actually came in from a longtime follower of yours. Underrated or overrated, RAL will step foot on a golf course before 2030.

[00:57:34] Raoul Pal: Well, there’s ways of. tricking me into getting onto a golf course, but no, I just look at those clubs. They make me shudder. Had

[00:57:43] Kevin Kelly: to keep them in purview.

[00:57:45] Raoul Pal: Friends of mine laugh that the last time I played golf, I just dumped my clubs in a bin and walked off and said, I’m never fucking doing this again. That was a,

[00:57:52] Kevin Kelly: it

[00:57:54] Raoul Pal: was like, I was never great at golf and he used to play it with mates.

And um, I was on a golf course and there was some old retired Dutch people behind me, tutting at me when I was hooking or slicing it. So shouldn’t be on the golf course. I’m like, fuck you. Somebody’s got to learn somewhere, but it’s golf is like, it’s the whole etiquette thing around golf that I just like.

I, I, I can’t do this. Yeah,

[00:58:19] Kevin Kelly: I hear you. I hear you. Yeah. There’s a, I’ve played with some of the most interesting, like greatest people when you get paired up and then there’s certain times you get paired up with people, you got people behind you that are just, I don’t know if they’re having a bad day or what, or that’s just their, their MO, but it can be frustrating.

I get you.

[00:58:35] Raoul Pal: Yeah. So anyway, I’m also shit at golf, so that doesn’t help. I’m just not the most coordinated person in the world.

[00:58:41] Kevin Kelly: Well, I’m going to keep my fingers crossed. Maybe I’ll be able to change that. Um, well, listen, as always, really appreciate you carving out the time. I think this is, I know, super valuable for me every time we talk.

I think it’ll be really valuable for, for the rest of our listeners and our audience really quickly before we go. People, I think a lot of people already familiar with you, but if they want to hear more from you, want to, you know, get more of your insights, read more of the stuff you put out, what’s the best way to find you?

[00:59:06] Raoul Pal: What is coming up in November? I write a service called The Exponentialist with David Matten. So it’s a, it’s kind of a spin out of some of the stuff I do at GMI, which is all around exponential age technology. I think we’re going to be running a free trial for all of October on Real Vision. So if people aren’t a Real Vision member, Right now you’re getting pro macro free trial.

So go to realvision. com. Just sign up for free. There’s a ton of stuff. There’s an amazing community. We’ve got, you know, an epic event coming up in Miami, but if you’re interested in this topic, I write a lot about it, but it’s in a service called the exponentialist and it’s going to be free for a month of November.

So I would just go and sign up now and enjoy it when it comes.

[00:59:45] Kevin Kelly: Yeah, I can vouch for that too, because I’m an avid reader of just about everything you put out. Um, and it’s all very much worth, worth the time and worth the money. So listen, sir, really appreciate you stopping by, taking the time.

[00:59:57] Raoul Pal: Yeah. I feel like we barely scratched the surface.

[00:59:59] Kevin Kelly: Yeah. We’ll have to do this again.

[01:00:01] Raoul Pal: Yeah.

[01:00:01] Kevin Kelly: I love it.

[01:00:02] Raoul Pal: All right, my friend.

[01:00:03] Kevin Kelly: All right. Take care, everybody.

Delphi Oracle

Latest on Delphi