2028年全球智能危机：AI会引发市场崩盘吗？

I think that if the mortgage market is in trouble, if there’s a credit crisis, there’s one thing you can be damn sure of: the U.S. Treasury and the Federal Reserve will gang up and start blasting money into the economy with a fire hose. They’re not just going to sit idly by on the sidelines while Rome burns, lose their jobs, and go down in history as the world’s worst economists, politicians, and stewards.

They will absolutely do what they’ve done every single time since 2008, which is just inundate the market with helicopter money. I think your bitcoins will be looking pretty good there. I think that’s the best passive hedge for that scenario.

That music was—it just hypes me up.

Dude, you’re good. You always hype me up with that. I mean, you are one good producer, Jonah.

Thank you.

I like that. I like the new thing that Blockworks did with the glitch at the end of the 1000x intro. That was dope.

Anyway, we love a good—

We love a good glitch, Jonah.

We love a good glitch. All right, so how do we—did you promote the livestream here? Because Twitter’s a little whack these days.

The answer seems like no. Once again, the livestream is absolutely not promoted. We love these issues.

1. How To Survive AI Disruption

Yeah. And I think our listeners love them, too. One thing that I’ll tell you is that we’re supposed to be so nervous about the future of AI, and the creator of xAI can’t even get their livestream technology right on Twitter. I don’t know how afraid we’re supposed to be of these AI engineers if things keep breaking with Twitter.

You tell me, Jonah. But I’m not so nervous about these guys right now.

2. Citrini’s 2028 Global Intelligence Crisis

I’m not so nervous either, Avi. I feel like there’s a lot of room for human intelligence as of yet. However, I did get a lot more bearish on the back of that Catrini Research piece. I did read the rebuttal as well, but I think the rebuttal has a key logical defect, which I think we should discuss, too.

Anyway, I don’t know. What are we doing here? Are we going to talk about audio, video issues, and Twitter sucking until we get more viewers, or should we just jump right into it?

I think we’ve got to jump right into it.

Yeah, me too. I think it’s important to point out that the last 24 hours have basically been entirely consumed by everyone freaking out about this Catrini article that people have been pushing out.

To summarize the arguments presented in the article, I’ll sum it up in one sentence: AI is coming to eat us all. We’ve heard this over and over. There’s been a lot of fearmongering about AI, and there have been a lot of realistic takes on what AI is going to do to disrupt our economy.

There have been sectors that have been hit, and we’re obviously seeing some pretty massive growth in the capabilities of some of these AI agents, with Claude Code and all these—the idea of the SaaS apocalypse, which so far has been true. You’re looking at a lot of these SaaS companies down a ton.

But Catrini’s article makes the case that it extends beyond software. This extends to white-collar work in general, across the board. What we’re going to experience over the next 3 years is a massive wave of unemployment because there’s a cyclical factor to what’s happening in AI: the more that companies invest in AI, the more that they kill themselves.

I think the core premise of the article rests on the idea that AI removes a tremendous amount of friction, and friction, in Catrini’s words, drives a tremendous amount of the economy. He makes the case that removing friction will remove the 2.5% to 3% interchange fee that cards produce. He makes the case that agents will be able to cancel all of our subscriptions that we don’t use.

He makes the case that it’s basically going to get rid of all of these desk jobs that exist for white-collar workers who don’t necessarily produce a ton for the companies. That means that a lot of these companies are just going to completely lose their moat because of the new advances that are coming in AI.

He makes an argument that DoorDash, for example, exists only because it’s an app on your home screen that you use to order food when you’re hungry. That’s the main value-add of DoorDash. In the future, when everyone vibe-codes their DoorDash equivalent and there are hundreds of different types of DoorDash, agents will simply choose the DoorDash that is the cheapest.

Brand is going to be abstracted by this idea that anyone can create software and anybody can compete. We’re going to end up in a world where the moats that exist for many companies that employ people go to zero. That’s really the main argument that he’s making: we’re going to end up in this spiral where all of these companies are going to end up going to zero, and they’re actually going to end up killing themselves by investing in AI—

—and then there’s a credit crisis.

That was your takeaway from the article? I’m curious—that’s how I interpreted it. How did you interpret it?

Yeah, I think that’s a really good summation. I’ve got a few things to add here. Basically, what he says is that there are a couple of follow-on effects there. As you mentioned, friction is what drives the services economy. The United States is a service economy.

Basically, what that means is that we have become accustomed to paying people and service providers to help us overcome annoyance and friction. Artificial intelligence will replace that. Instead of paying a real estate agent 5% to 7% to shop for a home, an agent will just download the entire MLS database and shop for a home on your behalf. You’ll pay a 1% commission to buy a home.

That puts an entire class of white-collar workers out of jobs. That happens inside companies. Salesforce basically says that a company that spends 95 million on people and 5 million on tech will spend 70 million on people and 20 million on AI. That’s just 90 million spent instead of 100 million. That spending basically leads to this deflationary crisis.

Highly paid SaaS employees earning 150K to 500K a year get fired and become Uber drivers earning 45K a year, except Uber also gets squeezed because the app on your phone isn’t Uber anymore—the app that you just go back to regardless of the price. It’s an agentic rideshare finder that will just ping the hundreds of new services that are all delivering cheaper marketplaces for local rideshares.

Everybody’s costs get compressed, and basically all the wealth accrues to a few owners of the means of AI production while everybody else is screwed. That leads to a credit crisis, which starts in private credit but then metastasizes into the sources of permanent capital, like insurance and reinsurance. Then there’s a lukewarm policy response, a mortgage crisis, and calls for redistribution and socialism.

The mortgage crisis, the way that he frames it, is that unlike 2008, when it was just bad borrowers signing up for mortgages that were defaulted on day one, now it’s prime borrowers—somebody with a 780 credit score and a 500K-a-year job at a SaaS company in San Francisco. That 500K is no longer recurring income. It goes to 45K when they start driving for Uber, cleaning toilets, and working 3 jobs to support themselves.

Prime borrowers default as well, as opposed to in 2008, when it was subprime borrowers. Once the housing market collapses, everything unravels. Then aliens invade the Earth and we all die. So that’s the doomer narrative.

I didn’t catch that part of the article, but maybe I didn’t read it closely enough.

Yeah, that’s sort of my take. I think it’s extremely compelling. There are a few different things that are so great about it. It takes the likely Luddite argument against AI and frames it in the form of stories that you can relate to. That’s why I think it’s going so viral and touching a nerve. It has these incredible zingers.

I could read you a couple:

“Over the past 50 years, the U.S. economy built a giant rent-extraction layer on top of human limitations. Things take time. Patience runs out. Brand familiarity substitutes for diligence. And most people are willing to accept a bad price to avoid more clicks. Trillions of dollars of enterprise value depended on those constraints persisting. It started out simple enough. Agents removed friction.”

I think the real zinger of the entire article—and then I’ll stop talking here and push it back to you—is this:

“It should have been clear all along that a single GPU cluster in North Dakota generating the output previously attributed to 10,000 white-collar workers in Midtown Manhattan is more economic pandemic than economic panacea. The velocity of money flatlined. The human-centric consumer economy, 70% of GDP at the time, withered. We probably could have figured this out sooner.”

And here’s the real line: “We could have figured this out sooner if we just asked how much money machines spend on discretionary goods.”

Hint, it's zero. Right. So that's kind of the crux of the article for me. Do you buy that? Do you buy the doomer spiral there? What was your take?

I think it's very compelling. I do think that oftentimes, when new technology is invented, there's sort of a gap because our economies are not instantaneous. Things don't happen in a perfect vacuum, right?

When you're modeling things out, sometimes, if you've ever taken a physics class, you'll say, “Assume no wind resistance,” or, “Assume this chicken flying through air is a perfect sphere.” That's what a lot of times people—economists—will try to do with the economy. Then, when you have to get a little bit more serious about modeling, you have to think about where these latencies come in, because the market is not incredibly efficient. Obviously, there are inefficiencies; otherwise, we wouldn't be here trying to trade all day.

The way that I think about this is that there are parts of this that are true, but the general principle that our economies and humans can't correct for these imbalances—I don't buy that. Throughout history, we've always figured out how to deal with new technologies. It's just that sometimes there are slight inefficiencies in figuring out how to actually deal with them.

For example, when the Industrial Revolution comes about, the tick-up in unemployment that everybody expected only happened a little bit. It didn't happen in aggregate, because what people did was move on to other jobs, right? There were new jobs that were produced and new things that actually happened.

When the car first came out, people were like, “Oh, wow. This is absolutely going to decimate the entire horse industry. It's going to destroy everything.” What actually ended up happening is that it completely decimated the entire industry dedicated to horses. Basically, everybody lost their jobs there, but they moved on because the cars opened up so many different types of new jobs that were needed.

Road construction went through the roof. Restaurants between cities that used to be far away started popping up like wildfire, because people could now take road trips in their cars. It unlocked an entire section of the economy that didn't necessarily exist before, right?

Is it possible now that all of these software engineers—all of these people who get laid off from their jobs at DoorDash, Google, and Meta, which are massively cost-cutting—just start building lifestyle businesses for themselves, now that it's so easy for them to build applications? Do they just start building lifestyle businesses for themselves? You get 10,000 people paying $10 a year for something that you build, which isn't that crazy in the world of social media.

Think of all the new massive applications that might pop up. Maybe there's an explosion in new, interesting software games, right? I think people always figure it out. At the end of the day, there's always an end state of the economy that ends up being stable with basically every technological advancement.

I think making the case that this is so wildly different is kind of tough. The other thing that I think is that it's very, very possible that, because this is the first time we've actually automated human thought and not just human physicality, maybe it does take a little bit longer to figure this out.

Maybe the Catrini thesis—the end state, is what I'm saying—does play out by 2028. But prior to individuals figuring out what to do and new jobs popping up, what you get is this massive unemployment wave. I don't know if I can really argue super hard against that.

The only argument that I have is that it's going to take a lot longer to implement these changes than Citrini thinks. It's going to be a while to change human behavior. Everybody has suddenly been interacting by clicking on an app, which people have been doing now for 20 years. Are you really going to change consumer behavior in a year and a half, by the end of 2028? Maybe 2 years? Are people really going to be only interacting with AI agents?

I think they will.

Are they really going to change their behavior that much?

Yes, I do. I think the way it's going to work is that some new app is going to come out where, instead of clicking on Uber, you click on the find-your-car app that queries all the different services. There won't just be Uber and Lyft anymore; there'll be 63 others that connect on the back end, don't create consumer-facing apps, but do have marketplaces that are connected to different taxi services.

I click on a shiny app that looks exactly like Uber, and it guarantees me a cheaper price than Uber. It shows me the Uber, Lyft, whatever prices, and shows me what I'm buying. I think somebody will build that on the back end. It'll be an agent, and it won't be a dedicated marketplace constructed by a Fortune 500 company.

I do think that's possible because the consumer basically just has to say, “All right, I download a new app and use it,” and on the back end, it's an agent. No, I don't think random people will be doing their own GitHub commits.

By the way, for the listeners, we're getting to the trading opportunities in a second, but we have to set the table with this stuff. I think it'll look like an app, but it'll be agentic on the back end. Why not, right? Why wouldn't somebody rather get cheaper prices every time than Uber and faster cars because they're pulling 16 services, or 70, or whatever it is?

Look, Lyft is almost always cheaper than Uber, but people still use Uber.

Yeah, but I think the point is that somebody will create—

No. I think that the AI doomers sort of talk out of both sides of their mouth.

No, you're missing the point.

They say, “Look, we're so early. This is just starting.” Meanwhile, 84% of people have never used an AI application, according to their own research that they're producing, and they're saying that in 2 years every single person is going to be using an agent. To me, it just doesn't pass the sniff test.

I do think in the future—in 5 to 10 years—absolutely, right?

But in a year, you're really going to make the argument that by the end of 2028 we're going to be in a full-on AI meltdown because consumers have changed a huge portion of their discretionary spending, and decision-making is done by people over the age of 40? You really think those people are going to change their behavior so much in the next year?

No. No, I don't. I mean, maybe Uber's a bad example, but the cost of building an Uber has gone down with Claude. That's true.

There's no debate there.

Other people will build, whether it's in the rideshare market that we're sitting here nitpicking on, or whether it's some other service-economy element. It's cheaper to build software, so people will build cheaper software. Then there's the switching cost.

What the Catrini article argues is that, eventually, the back end that does the wiring will be cheaper to build, and the front end will be able to point at more back ends than the current vertically integrated, relatively limited set of choices that people have across the economy and various sectors right now. That sort of all-to-all marketplace-type effect will lead to deflation and job loss.

I kind of buy that. At first, when this AI thing happened, I was kind of dismissive of the doomers. I was like, “This is light bullshit. It happens. You know...” But I read Fooled by Randomness like everybody else earlier in my career, and the Nassim Taleb argument—it sucks that he's an antisemite now, but the Nassim Taleb argument that the turkey thinks everything is fine the day before Thanksgiving because it's never had its head chopped off before in its life—is compelling.

You do have to ask yourself, “Is this time different?” The Catrini article presents it in a very well-worded, compelling way: The reason why the likely Luddites were always wrong is that human intellect was the bottleneck that allowed the physical things that technology abstracted away from human physical limitations to be filled in with new jobs—new knowledge-work jobs. Now that knowledge work is being abstracted away by code, I'm struggling to debate the likely Luddites at this point.

Catrini has basically laid it out. Where I think he gets it wrong is that he says the policy response will be political. I think the policy response will be monetary. If the mortgage market is in trouble, if there's a credit crisis, there's one thing you can be damn sure of: The U.S. Treasury and the Federal Reserve will gang up and start blasting money into the economy with a fire hose to prevent it.

They're not just going to sit idly by on the sidelines while Rome burns and they all lose their jobs and go down in history as the world's worst economists and politicians and stewards.

[Speaker?]

Like, they will absolutely do what they've done every single time since 2008, which is just inundate the market with helicopter money, and I think your bitcoins will be looking pretty good there. I think that's the best passive hedge for that scenario.

Now, Bitcoin may go down if people are getting fired from $500K-a-year jobs and cleaning toilets for $20K a year until they get fired by Optimus, which cleans the toilets faster and faster and better, and then they're living in tent camps. There is some real doomerism here that I feel like is just hard to imagine. It's so bearish that it's hard to imagine.

Needless to say, the helicopter money will come. That we can be sure of. The other thing I wanted to touch on—maybe I should stop talking for a second and let you respond to all that—but I do want to touch on the rebuttal. The rebuttal—I wanted it to be right, but there was a key hole in the logic, and it was mega. So I don't know, what should we do? Should we talk about the rebuttal? What do you think of what I just said?

Go ahead. Talk about the rebuttal and what you thought the flaw was. Then I think there are some arguments that we need to address here. The core thing that I want you, the listener, to keep in mind is that the argument that I'm making against this idea is not that our economy is going to stay the same. Our economy is going to be radically transformed by AI. The jobs that people hold in the next 2 years are going to be very different from the jobs that people hold today. There are going to be massive layoffs of people at tech companies. There's going to be a massive drawdown in terms of cost. All of these things Catrini and I agree on.

What I disagree with is the idea that there is no cycle. The argument that he makes is that unlike other economic phenomena, where the cycle is: this becomes more efficient, layoffs happen, and then people find other jobs because money is freed up to invest in other areas, or costs come down, and then everything ends up at a different, stable equilibrium. He's saying that's not going to happen this time. That cycle is actually a doom loop. That stabilizing cycle of capitalism that has happened over the last 350 years is a doom loop because we finally automated human intelligence. And that's really the key point that I want to disagree with—not that our economy will be radically changed. So that's what I want to keep in mind.

Then I think you go into the rebuttal, and then we can talk about that. And then after we talk about this, guys, I'm going to go into telling you: because our economy is going to be radically transformed, what do I think, and what does Jonah think, the best potential investment opportunities for you right now are?

3. The Bull Case For AI & The Economy

Yeah, agreed. So we're getting to the trading opportunities, because there are some that, whether the rebuttal is right or whether Catrini is right, would perform really well in either scenario. If you're willing to take a side, then you can do extra well on certain trades.

The rebuttal does not disagree with Citrini on a number of points. The rebuttal agrees there's going to be tremendous job displacement. But where the rebuttal takes issue is with that. It says, and I quote, “The bears made an assumption that turned out to be wrong: that displaced white-collar workers would remain displaced.”

So Catrini says, “These white-collar workers, once they're kicked out of Salesforce and then kicked out of Uber and then kicked out of cleaning toilets, they're just permanently displaced. They're screwed.” The rebuttal says, “No.” And then it paints a scenario. It says, “What happened instead was faster and messier than anyone predicted. The same AI tools that eliminated certain roles also made it dramatically cheaper to start things. The cost of launching a business—software, legal, accounting, marketing, design—fell by 70% to 80% in 18 months. New business formation exploded,” blah, blah, blah.

So that's one piece of the rebuttal. The other piece of the rebuttal says, hey, just because real estate fees go from 5% to 7% of the transaction value to a few basis points because AI is helping the shopper analyze MLS, identify opportunity, and know fair value, that doesn't mean that we shouldn't focus on real estate agents—poor real estate agents, the women of Selling Sunset losing their jobs and not buying high heels anymore—and the economy going into a doom loop. We should ask ourselves, hey, what about all those homebuyers that now have an extra 5% to 7% of their entire real estate value in their pocket? Or the seller, or whoever the fee comes from—I forget. Are they just going to pocket that and stash it under a mattress? No, they're going to go and spend it. They're going to spend it on furniture, fittings, and finishings, and all the stuff that goes into their home. Or maybe they'll buy a nicer, more expensive home. So everything's going to go up, not down.

That piece of the rebuttal holds a lot of water. In the world that we live in, especially where AI is melting our brains into the idea that we need to buy this or that or keep up with the Joneses, I agree that every penny saved goes straight back into consumption. So that piece of the rebuttal is very powerful.

The other piece of the rebuttal, where it says that displaced white-collar workers would remain displaced, is just not going to happen. The idea that suddenly all of these displaced workers are going to create amazing businesses for free and that those businesses are going to thrive is flawed. And this is why, sadly, I did not find myself—even though I wanted to very badly—really vibing with the rebuttal.

It's like, I'm sorry, I don't care how easy it is for AI to help you pursue your passion of quilting after you got laid off from Meta. Nobody's going to buy your freaking quilts because they suck, right? The only thing—it's all relative. There aren't a million or a billion solo actors who just got laid off from middle-manager jobs to be entrepreneurs, solo entrepreneurs, selling products.

Eventually, I guess, maybe to our earlier argument, if agents can just pick products for you, then maybe there's no need to market your stuff, and you just get discovered by the robots if you build something good. But I ultimately think that, even much like human intelligence is today's bottleneck that's just been busted open by AI, AI-assisted intelligence is still a bottleneck, right? There will always be a best product. In the future, it will be AI-assisted; today, it's mostly human-engineered. But there will always be a bottleneck, is what I'm saying. There isn't room for everybody to succeed with AI. Some people will use AI better than others, and so all those laid-off tech workers—there will be a ranking from best to worst. They will not suddenly all succeed at the same time because of AI.

So, unfortunately, that is such a massive logical hole in the argument that I think we're in for some real turbulence. I don't know. What do you think of that?

Look, I think that was kind of the argument that I was making earlier, which is, I do think that there will be—here's what I think is going to happen, Jonah. Regardless of whether those people end up starting companies or not, they're going to end up somewhere. The most likely place for these people to end up is in jobs that AI cannot effectively bring down the cost of. So they're probably going to end up either in blue-collar jobs. Maybe, as you said, as the Citrini article itself said, maybe where they end up is driving Uber, or maybe they end up being the delivery people for DoorDash, which still, as of now, requires physical delivery. Maybe in the future they'll be displaced by drones too.

But drones—

Basically, where they're going to end up is in the sectors of society that AI cannot effectively bring down the cost of. So the argument is kind of missing a logical piece here, which is: if you think that AI is going to come in and massively reduce friction in consulting, massively reduce friction in payments, massively reduce friction in software, and massively bring down costs for all these companies to the point where, to be competitive, they're going to have to bring down the cost of their product in a huge way, then that means everything that AI touches—the cost of those products—will go down.

And then what are you left with? You're left with things like food. You're left with things like physical items, furniture, all of this other stuff that maybe requires more actual human labor. Art, for example, right? Maybe now these people go into art, or they're going to end up in areas that AI cannot touch.

What areas? This is what I'm getting at. What can't AI do? Humanoid robots will be better than humans at lifting boxes and driving forklifts.

Well, let's see how long that takes. But basically, the argument is that AI is going to touch everything—literally everything—and leave zero jobs for humans.

I don't necessarily believe there will always be jobs. Something will be inefficient somewhere, and let's say government—that's not going to be done by AI.

But fine, let's say, in general, that yes, even if AI literally touches everything, guess what? The cost of everything is going to collapse then. So, if people aren't making a ton of money but the cost of everything collapses, where do people net out?

If you end up taking an 80% pay cut, but now the costs of everything are down 70%, that isn't as catastrophic as people might think. What I'm saying is that regardless of what happens—even if people go unemployed and have to go to the government—the end result of AI is simply streamlining and making our economy more efficient, freeing up time and capital to reallocate to the areas where at least we can both agree that in 2 years AI won't touch everything. In 4 years, AI won't touch everything. There's a progression here.

So, what happens is this particular scenario that we're talking about, where AI nukes everything by 2028 and unemployment goes up: those people are going to add to the labor supply for the areas in 2 years that AI isn't touching. That's where they're going to end up. They're going to increase the supply of labor in the areas that AI isn't touching.

Basically, what I'm saying is that there is a cycle here. They're going to bring down the cost for the things that AI can't touch, and the things that AI is touching—the cost is coming down on.

Okay. So we're getting to the trades, but I want to talk about that. You bring up a very important point, which is AI creates deflation.

Now, the big question mark is: do wages deflate more than goods and services, or do wages deflate less than goods and services? The bear scenario—the Citrini argument—is that wages deflate more than goods and services, so real income goes down for most people.

The rebuttal should have said, “Hey, everything's going to deflate.” They shouldn't have tried to say everybody's going to become their own successful entrepreneur. That was BS. That was not convincing. What the rebuttal could have said is, “Hey, AI is going to deflate goods and services more than wages deflate, so everybody's better off, and we'll just, you know, a dollar will suddenly be worth a lot more again.”

Now, here's where you're coming in on that: people will just move from places that AI automates things to places where AI doesn't automate things. But that's a diminishing battle that I don't think humanity can win, because unlike previous disruptions, where people can retrain at the speed of people, right now I think—let's say AI eats a third of the economy in 2 years, another third 2 years after that, and then 50% of what's left 2 years after that—people just can't adapt and move fast enough. They'll just be running away from Skynet, and that's not a recipe for economic health.

The other problem that you didn't touch on in your argument, which I think leads to our first mega-trade section on the 1000x podcast, is that we all agree there's going to be deflation in wages and goods and services. Everything's going to get cheaper. The rebuttal says it. The Catrini article says it. We're all on the same page about deflation.

You know who deflation screws? It screws people who owe money, right? It screws borrowers. So, if you have a mortgage and everything deflates, you don't really care that you can go to the dollar store again and buy a bunch of stuff—a bunch of cheap Chinese crap—for a dollar that currently costs $10. You care about your mortgage payment that you can't afford because your salary went down.

So, the government is going to have to do one thing, which is print to stabilize, to counteract deflation.

Well, or what basically they have to do is, when faced with deflation—if we get deflationary prints—rates are going to zero immediately and everyone is refinancing.

That's really what's going to happen. Right now, we're still in an inflationary environment because of all the money that the government has pumped into the system. I think that's one of the myriad of reasons that we've discussed on this podcast why gold has done well. Our rates are high right now; we have so much room to cut relative to history. The moment that we see deflation in this economy, rates are going to zero.

4. AI Trading Opportunities

That's what I just said, right? People are borrowing money. So, what does well when rates go to zero, Jonah? Let's start there. What trades—if you believe in the Catrini thesis, if you believe in the growth of AI, and if you believe in deflation and rates go to zero—what are we buying?

Because, importantly, this is one area where we both agree: even if the economy doesn't collapse, we both agree here that rates have to go to zero.

So, what happens when rates go to zero? I've got my answers. I'm curious about yours.

Well, there's the obvious bonds trade, but that doesn't give enough juice for the degenerate listeners.

You could lever it up.

You could lever it up. And then there's the obvious crypto trade. There's the obvious gold and precious-metals trade, and there's the obvious equities trade.

Basically, what you should be looking for as a listener is to be short human-output businesses—which is obvious, your point on AI in previous podcasts—and long inflation-linked cash flows.

I'm an LA guy through and through. Now, I would want to be long—my buddy runs Miramax. And now you're probably asking yourself, “Hey, what do you mean your buddy runs Miramax?”

What do you mean your buddy runs Miramax?

It's a private-equity-owned asset now. Harvey Weinstein is in jail, and it's not an active movie studio, but Pulp Fiction—people keep renting it, right? So, that's an inflation-linked trade.

That makes sense. One trade that I've been kicking around because of this—I mean, let's say we both agree, even if we don't necessarily agree that the economy is going to totally collapse because of this and unemployment is going to go to 10%, which I actually think is very unlikely.

I think it's unlikely too.

I think this is a very unlikely scenario, just because of everything that I've discussed. I think the slower pace of AI growth is one of the reasons that there's going to be a slower pace of AI growth: there's a physical constraint to data centers being built, there's a physical constraint on chips, and there's a massive physical constraint on energy right now.

So, I'm very, very, very bullish on energy because of this. I'm very bullish on uranium, which I've talked about, because of this. And this plays into the zero-rate thesis as well, but utility companies are going to end up making a ton of money.

One, because they give out dividends, and if rates go to zero and their dividends stay at 3% to 4%, which seems like where they are right now, people are going to crowd into those stocks too, because there's tremendous demand from data centers for energy, right?

Be more specific on your energy bull thesis. Energy means so much to me as a commodities guy. Are you bullish on coal? Are you bullish on European natty? What are you bullish on specifically? I want to know.

I'm bullish utilities.

Okay.

In this specific environment, go out there. You don't necessarily want to be tied to one region, as you saw with California. You can get totally nuked on that.

But if you buy a basket, I'm bullish on your Exxons of the world. I'm bullish on your Chevrons of the world. There's going to be much higher demand for energy in this particular environment, and I'm very bullish on uranium because I think we're going to start building nuclear reactors again in the next 5 years, and we're going to pull forward some of that.

Maybe you want to go out there and buy Constellation Energy, for example, right? They're—

I'm bullish on energy-trading businesses. So—

Very—that's also true. I think energy-trading businesses, if they—I don't know if any of them trade publicly that you could go invest in. But I think they're going to absolutely clean up here.

I think so too. I'm bullish on anything. Basically, what I'm expecting—I just want to refine a statement I made earlier. I think the Catrini thesis was far more compelling than the rebuttal because the rebuttal contained the logical hole previously discussed.

What I will say is, I do not think that this will lead to economic collapse. I think the path is important here. All of this is path-dependent. I think what will happen is that the Citrini thesis will begin to play out, and then what Citrini didn't really talk about that much is what we just touched on, which is that there's going to be a fire hose of cash pointed at everybody who's sort of disadvantaged by this.

That will lead to inflation-like trades, but more importantly, volatility. Volatility is the word of the day. It's like Sesame Street here. Basically, the word of the day is volatility. Anything that benefits from volatility, that is long optionality at cheap levels—so that's trading businesses, utilities, and certain private equity plays—should benefit.

I think Bitcoin is an antifragile long-vol asset that you can kind of just be long by buying some tokens. I think we're in for a wild ride here. If you could get your hands on some Poly Market stock or Hyperliquid tokens, anything that benefits from just the soup getting churned around and that doesn't benefit—or, in fact, suffers—from environments where nothing happens, that's what you want to own.

You want to go back to my unfortunate Taleb line of thinking here. You want to own the antifragile assets that benefit from volatility. And I can think of a million of those. Basically, it's time to start buying options in a way where you can survive and not bleed out.

Look, I can tell you one reason why this world is going to be a lot more volatile just from observation. Catrini puts out this article last night—or when is it? Yes, it was literally yesterday. He puts out this article, and I know we've spent a lot of time on this podcast. I've personally spent a lot of time attacking this article.

It's a really good article, and I think he put a lot of thought into it. I really respect him for going out there and pushing this thesis because I think it's tough to be a bear, especially publicly. People, I think, are very uncomfortable with the idea of a bear thesis, but people loved it and it went ridiculously viral.

Not only did it go ridiculously viral, every stock that he mentioned in his article is down today, Bloomberg is reporting on it as the reason for the crash, and all the PMs are talking about it. I think people, from a psychology standpoint, are still trying to call the top on this, right? People are trying to call the top on this bull market. Bull markets don't end until nobody's trying to call the top.

A bull market ends when an article like this gets no traction. I'm sorry, but this getting 20 million views in and of itself is bullish to me because it means that people want to be sidelined, and this is happening.

The article even calls for it. It says it's going to melt up before it melts down.

Yes. This article itself comes out after everything in the market is decimated. All of these SaaS businesses are totally decimated. Google is 10% off the highs. Actually, the only tech stock that's doing well is Apple, which I actually think—I'm getting very, very, very bullish on Apple, and we can talk about that.

Me too.

In a bit. It's Apple, X, XLE, and URRA. I've obviously still been bullish on gold. Maybe you want to go buy some TLT.

Samsung too.

TLT has been doing well. Maybe you want to lever up that trade based on everything that we've talked about. And also, I think soon we're going to get some good entries into some of these European defense stocks because I think that's a megatrend.

I do respect Catrini a lot, but this response to this article could not have come 2 months ago, when everything was at the highs, right? The market is clearly very, very, very nervous right now. And also, more than that, if an article like this is able to move the markets, it means that PMs and big money and hedge funds don't have strong theses right now.

It means that they're moving around money. They're trading like crazy, and that's going to lead to a lot of volatility because people aren't sticking in positions right now. People are very nervous. People are very nervous because we are on the precipice of great change.

There is a lot that's going to happen over the next 5 years that is going to create a tremendous amount of trading opportunity out there, a tremendous amount of volatility, and nobody knows what the hell is exactly going to happen. They're trying to predict what's going to happen over the next 3 to 6 months right now.

We've moved from where, 10 years ago, hedge funds were sitting in positions trying to predict the next 2 to 5 years. Right now, everyone's trying to predict the next 3 to 6 months. And that's why I think the timeline has really compressed, because we're in a period of such radical change.

Anxiety.

You can't push it out that far. And so, people are really getting in and out of positions really quickly, which for you, the trader, makes this maybe the best time ever to trade. If hedge funds that control billions and billions and billions and billions of dollars are moving on the order of 3 to 6 months, you can front-run those trades super easily because they're actually getting in and out of these trades pretty quickly now.

You can just ride trades with them, right? Because they're going to work into trades over a week. It's going to take them a little bit longer. So, I'll give you a great example of this. If a hedge fund gets very, very, very bearish on SaaS, it's going to take them 3 to 4 days to get all of their positions on.

It's going to take you, the retail guy, about 3 to 4 seconds—one click of a button. So, you actually have an advantage here. You're playing the same game, but you have an advantage. You just have to be paying attention.

You have to be watching this podcast because maybe what happens is, in 2 months from now, every hedge fund in the world is going the long-bond trade because they're getting nervous about deflation. And guess what? You listened to this podcast; you got in there a little bit early.

Yeah. I mean, another example of how you can beat the hedge funds here: I'm crazy bullish on Apple. I could talk about why. I was glad that you mentioned it too. It's crazy.

It's down 1.5% today on a Catrini-article-led sell-off. The Catrini article literally says that the stock market melts up before it melts down as companies become more efficient using AI and fewer people, right? The rebuttal makes a good point, which is that companies, like people, will pursue new opportunities with the cash they save.

They won't just pocket it and not use it and have it become stranded assets. They'll either do share buybacks or pursue higher-ROI opportunities than what their cash was previously tied up for, which was people who have been replaced by AI. But another thing the Citrini article says is that computing's going to the edge.

That is a theme that I'm betting on personally. I'm kind of all in on that. Computing going to the edge means token usage per person goes from 0, 3 years ago, to 10,000,000 tokens a day, whatever the article says. I'm not an expert on this, but basically, every person uses more AI and more compute per day.

And it does say that a lot of that won't take place in the data center. It'll take place on the edge. So, your edge devices will get smarter. They'll consume more, produce more compute, and consume more energy. Apple is the prime beneficiary of that.

They have the luxury device that's in everybody's pockets. Models will be run inside your pocket pretty soon, as you need them, whether you realize it or not. And the obvious number 1 luxury-device maker in human history is going to benefit from that megatrend, but it's getting whacked today because a bunch of hedge fund professionals are like, “Oh, shit, I got to sell because there's a doomer narrative taking place, and I need to momentum-trade that.”

So, this is your time as a longer-time-horizon retail trader to scoop Apple as it's coming off and hold on to it for longer than the time horizon of the guy who's puking it, which is probably a few weeks to a few months. You ride this thing into literally what the Catrini article suggests, which is mid-2027, at which point you're probably up 50% with very little downside on something that's already totally blue chip—no pun intended here.

So, basically, opportunities like that let you have a longer view and buy things like this. This is like Amazon stock selling off because there's a tech crisis in 2000.

Yeah. I mean, another example of how you can beat the hedge funds here is that Anthropic says—there's this headline that came out 24 minutes ago. Kind of funny.

Anthropic says cloud code can automate Cobalt modernization.

Automate what? Sorry.

COBOL modernization. Cobalt is what a lot of financial applications are built on. It's a lot of the old—you know what COBOL is? It's old.

Yeah, it's one of those old programming languages.

Very old programming languages that only boomers know about, like Jonah. It's the backbone of a lot of financial products, actually, funny enough, because they were built a long time ago. IBM does a lot of consulting to update these code bases, and Dario Amodei, I guess, comes out and says, “We're going to take that business away.” They don't mention IBM. I want you to guess how much IBM is down since I had left. Don't look.

I won't. 4%.

It is down 10%.

Oh—

Suck Business Machines.

Sorry.

It is down from peak to trough 7.5%.

Did you just say “trow” and not “trough”?

Yeah, peak to trough.

Okay. I was like, maybe I've been doing it wrong my whole life.

I actually have no idea. I just feel like Candace Owens on that one. Peak to trough.

Yeah, it's kind of crazy. I think I would not be long International Business Machines. That's basically a human consultant. It's like, be short the human output. It's like short, short.

Actually, funny. That's a good point. I know it's down 10%, and I hate to say it, but basically on a bounce—if you get a bounce from IBM—maybe you do a little pair trade here. Consider getting really long Apple and getting really short IBM.

IBM's the new WIF on an IBM bounce because IBM looks pretty freaking nasty. I'm not going to lie. I mean, what is their business? If they're a consulting business and they're a software business, what are they going to do, right? But maybe what you want to do is start funding some IBM engineers to go start their new companies.

Jonah, it's funny: There's the old saying in business, “Nobody ever got fired for choosing IBM.” I think that's over, right? Now you could literally get fired by somebody for being like, “Sorry, you spent how many millions of dollars paying IBM to do something instead of just doing it yourself?” It's a bit ridiculous.

Actually, funny. That's a good point.

I think the adage has flipped, and what's so interesting about this whole market is it feels like we're in the early innings of a paradigm shift. We'll get plenty of insane volatility. The best way to make money from trading is getting a good entry point, right?

I think this is more of a trader's market. Publicly available markets are going to be more of a trader's market than they've been, basically, since the financial crisis and COVID. Instead of it just being this smooth uptrend, I think we're going to get wild volatility as AI rearranges things.

But again, I'm going to present an image here that I have saved on my desktop from South Park. What will not happen? This image is a meme. South Park basically coined this meme, which says, “Phase 1: collect underpants; Phase 2: question mark; Phase 3: profit.”

I think this is a lot of people's AI life strategy: Just come up with something, get fired from Salesforce, use AI agents to collect underpants, and then the profit will come. This is not going to happen. You have to be a little bit more thoughtful about how you approach this market than that. Sadly, this is all too common these days.

Yeah, this is great. Dude, let's stay safe out there. Avoid Skynet.

Avoid Skynet. Avoid IBM. Avoid computation between sunset on Friday and sunset on Saturday. That is another recommendation.

Oh, this reminds me. Before we close, I just want to say you said something earlier in the podcast that I want to touch upon. I'm not sure I agree with it, but I want to talk to you about it. You said that humans always figure it out, right?

Technological advances, societal upheaval—people always overcome. Even though we talked about a lot of doomerism today, I tend to agree with you. However, I would modify that statement. I would say some people figure it out. Others blame the Jews, and their society implodes, right?

That's how a lot of people deal with drastic winds of change. It's not entirely clear whether America will figure this out. Some people somewhere will figure it out. I believe in this country, and I believe that America will figure it out. But that's not to say that everybody globally is going to figure it out.

India's outsourced IT service economy may indeed tank, and we may see some issues in places that don't adopt AI intelligently or revert to hard-line communism as a means of redistributing AI wealth, or AI problems, or whatever.

So basically, in addition to all these trades, having skin in the game, having P&L fluctuating in your face, and knowing that you're emotionally attached is your best finger on the pulse, your best ear to the ground, your best telemetry into whether the market that you're invested in is going to survive and thrive or get mulched.

I feel like I have to take more risk now in order to peel back the layers of the onion and not just sit on the sidelines in analysis paralysis, but actually feel the changes as they happen. I would encourage everybody else on the call here to try and do that, just to actually sense whether you're in a winning position wherever you live and whatever you're long, or whether you're about to have the rug pulled.

I think that's a really good note, Jonah. I think we can leave it there.