The AI End Game: Boom to Bust? with Ed Zitron

Summary

---

Why Is This Happening? with Chris Hayes

Episode: The AI End Game: Boom to Bust? with Ed Zitron

Date: June 9, 2026

---

Episode Overview

In this episode, Chris Hayes explores a contrarian viewpoint amidst the current AI boom: Is the hype sustainable, or is the tech world building up to a bust? Guest Ed Zitron, tech industry critic, podcast host, and author, makes the case for deep skepticism about the economics, business model, and promised value of generative AI. Drawing comparisons to previous tech and financial bubbles, Zitron and Hayes break down how money is moving through the AI ecosystem, who’s making (and losing) money, and what would have to be true for the AI "end game" to avoid collapse.

---

Key Discussion Points & Insights

1. AI Hype in Historical Context

• Comparisons to Past Bubbles:
  Chris Hayes compares the current AI boom to the housing bubble and tech bubbles like the Metaverse and NFTs, where the logic of endless growth led to catastrophic busts.
  • "It's always a fine line between crank and profit." (Chris Hayes, 03:05)
• "Growth at All Costs" Mindset:
  Zitron argues that Silicon Valley's obsession with “growth at all costs” (the "grot economy") has run out of new real engines of hypergrowth.
  • "The AI bubble is a symptom of a larger problem with the software industry. The hypergrowth era is ending..." (Ed Zitron, 04:13)

2. AI: Is There a Real Use Case?

• AI vs Past Fads:
  Chris notes that, unlike the Metaverse or crypto, LLMs have plausible use cases (e.g., document review, coding), but efficacy and reliability are deeply debated.
  • "To me, the big difference is no one could ever really explain to me the use case of the metaverse… I don't feel like that way with AI." (Chris Hayes, 08:00)
• Reliability and Hallucinations:
  Zitron emphasizes that hallucinations—AI generating false or nonsensical outputs—are inherent and unsolved.
  • "OpenAI's own research says hallucinations are a part of these things. They're never going away." (Ed Zitron, 09:21)
  • Even with source gating, hallucinations cannot be eliminated, undermining trust for critical applications like law or coding. (09:55–11:10)

3. Tech vs Economics: The Investment Disconnect

• Historical Parallels and Their Limits:
  Unlike railroads or the internet, AI lacks a clear development or cost-reduction roadmap. With physical tech, scaling drove utility, but for LLMs, cost curves and utility gains are unproven.
  • "No such path exists for large language models. For that example to make sense, you would have to have a way in which the cost came down and the hallucinations went away. Neither of those appear to be happening." (Ed Zitron, 12:54)

4. Where the Money Goes—and Who Makes It

• AI Ecosystem: A Money Map:

  • ~$1 trillion invested into AI companies, data centers, and chips, mostly from venture capital, private equity, and big tech.
  • The "shovel sellers" (chip makers like Nvidia) are the only clear winners.
  • The main AI model labs (OpenAI, Anthropic) aren’t profitable; they burn cash to train and run models, especially for compute-heavy applications like code generation.
  • Notably, Microsoft and Google have shifted from using only cash flow to taking on debt for AI CapEx. (16:49–17:05)
  • "None of these businesses are profitable, not a single one of them. What's really interesting is none of them talk about the AI revenue." (Ed Zitron, 17:05)

  Notable Quote:

  • "When nobody wants to talk about the money and nobody can really precisely describe the outcomes, that's when people should get a little concerned." (Ed Zitron, 01:01, repeated at 17:37)

5. The Flawed Business Models & Subsidies

• Subscription Economics:

  • User subscriptions are massively subsidized—the cost to serve heavy users is far higher than the subscription price (e.g., $13 spent for every $1 received in subscription revenue). (21:27–24:37)
  • Hopes of raising prices to sustainable levels seem unrealistic, given the extreme costs and unproven willingness to pay.

  Quote:

  • “Their only customers appear to be themselves or very small amount of AI companies, all of whom are terribly unprofitable.” (Ed Zitron, 31:24)

• Data Center Explosion—and Risk:

  • Huge sums have been sunk into new data centers, mostly on debt. Many data centers are being built with the hope of future demand for AI compute which may never materialize.
  • Many are already struggling—rumored to be losing millions daily. (40:50–40:59)

6. The Dangers of Obsolescence and Debt

• Hardware Depreciation:

  • New GPU generations are released annually, but data centers take years to construct—meaning many will open already obsolete.
  • "You'll have an entire data center full of obsolete GPUs." (Ed Zitron, 39:54)

• Huge Debt Risks:

  • $178.5 billion in data center credit deals in 2023, yet less than $1B in associated compute rental revenue outside big hyperscalers and AI labs.
  • Should demand fail to meet expectations, a cascade of defaults and implosions is possible—the first cracks will likely appear in the private credit/debt markets. (32:09–32:46, 52:53–53:49)

  Notable Quote:

  • “Everything that's happening is a stress test of debt and equity. How much can venture capital spend? How much can hyperscalers afford? How much money is left in the coffers?” (Ed Zitron, 51:29)

7. The Only Profitable Player: Nvidia

• Nvidia’s Unique Position:

  • Nvidia profits handily from selling the “shovels,” i.e., the GPUs, while the “gold miners” (AI companies) hemorrhage money.
  • Nvidia is now even buying equity in AI infrastructure companies to keep the flywheel spinning.

  Quote:

  • "I don't think it's a good sign that the shovel seller is also paying for the digs." (Ed Zitron, 27:32)

8. The End Game: Cataclysm or Transformation?

• Lose-Lose Outcomes:

  • Either the bubble bursts financially, or, if the technology fulfills its most transformative promise, it could render tens of millions of jobs obsolete (and the societal consequences would be cataclysmic).
  • "There's no way out of some kind of cataclysm for this reason. Either your case is correct, in which case it's just not going to be profitable and the whole thing is going to collapse in on itself... Or...they can replace the labor of tens of millions of people. That's another cataclysm." (Chris Hayes, 48:58)

• Skepticism of AGI Promises:

  • Zitron argues that predictions of true AGI are built on extrapolating past tech trends without real causal basis; current AI is nowhere close to replacing human judgment or complex labor.

• Social Backlash & Boss Fantasies:

  • A coming "social contagion" as ordinary workers witness elites eagerly rooting for their replacement.
  • "Regular people have seen how their bosses think of them… You weren't in crypto. Also, I'm replacing you with AI." (Ed Zitron, 51:31)

9. Where Will the Crack Happen First?

• Zitron’s “Three Horsemen” to Watch:
  1. A data center construction project collapses before completion.
  2. An operational data center shuts down due to inability to pay debt.
  3. Accelerating defaults/loan distress in the private credit/venture capital sector. (53:54–54:20)

---

Memorable Quotes & Moments

• On Benchmarking AI Models:
  "They actually found that one of the anthropic models had just started going and looking for the solution, wasn't trying to solve it, just went on GitHub and did it... No, you asked a thing to do a thing and it did a thing.”
  (Ed Zitron, 14:08)

• On Software's Growth Ceiling:
  "All of these investments...were acquired or invested in based on the idea that they would grow like they did between 2005 and 2018. Now they grow like after 2018 there’s less money and there's only so many software companies you can build, only so many people you can sell to.”
  (Ed Zitron, 55:31)

• On AI as Utility:
  "Sometimes they'll talk about their own vision of artificial intelligence being like a utility, like electricity or water...utility companies are not super profitable. In fact, they’re regulated."
  (Chris Hayes, 55:52)

• On the Basic Truth of Bubbles:
  “Whatever financialization happens... in the end you got to make money. Things have to be useful and make money for everything to keep going.”
  (Chris Hayes, 55:09)

---

Important Timestamps

• 03:55 — Ed Zitron introduces his AI skepticism and "growth at all costs" critique.
• 09:05–11:10 — Discussion of AI hallucinations and limitations in coding/research.
• 12:54 — Comparison to railroads/internet and why those analogies may not fit AI.
• 17:05 — The profitability black hole and lack of real AI revenue from the big players.
• 21:27–24:37 — Subsidized pricing of AI, unsustainable economics of subscriptions.
• 27:32 — The "shovel seller is also paying for the digs": Nvidia’s entangled investments.
• 31:24–32:46 — The real AI compute companies aren't making money; the credit treadmill.
• 39:54–40:47 — Hardware obsolescence and data center construction lag.
• 48:58 — “No way out of some kind of cataclysm” dilemma.
• 52:53–54:17 — Where the first cracks will appear: loan defaults and data center failures.
• 55:09 — The hard reality: "in the end you got to make money."
• 56:25–58:03 — The possibility—and danger—of pensions/insurance funds being exposed to AI infrastructure debt and the implications if flagship projects like OpenAI fail.

---

Tone and Speaker Dynamics

• Chris Hayes: Inquisitive, probing, insists on dissecting differences between "tech works" and "business works"; often seeking precise economic and technical distinctions.
• Ed Zitron: Witty, cynical, and direct; relentless about the lack of profitability and skeptical of the industry's historical analogies and justifications.

---

Takeaways

• The AI gold rush is built on shaky business economics, with enormous risk radiating outward from a small set of profitless lab companies, hyperscalers amassing debt, and a bubble of debt-fueled physical infrastructure.
• Nvidia alone is reliably making money—everyone else is betting on an as-yet-unproven future where AI creates trillions in new value, a trajectory with little precedent.
• The first cracks will appear in debt markets, likely with unfinished or shuttered data centers, before potentially triggering a broader software or tech sector reckoning.
• If the bullish case wins and AI does revolutionize productivity, the scale of labor displacement could itself be socially and economically destabilizing.
• Beneath the surface optimism is substantial social anxiety and skepticism about whether AI is, in fact, delivering on its promises or simply setting up the next bust.

---

This summary distills the major themes, arguments, and exchanges of the episode. If you want a vivid, sometimes biting, and highly detailed takedown of the AI hype cycle, Ed Zitron delivers, while Chris Hayes frames and challenges the pessimism, ensuring a lively debate for anyone interested in technology’s future and its financial underpinnings.

Transcript

[JP Morgan Advertiser]

Managing your working capital with fragmented platforms makes it hard to optimize your cash flow. JP Morgan's Working Capital Accelerator brings payables and receivables financing together into one centralized platform to help unlock your working capital. JP Morgan 2026 JPMorgan Chase Co. All rights reserved. JPMorgan Chase Bank NA Member FDIC deposits held in non US branches are not FDIC insured. Non deposit products are not FDIC insured. Not all products and services are available in all geographical areas. See jpmorgan.com paymentsdisclosure watch disclosure day at

[Disclosure Day Advertiser]

home now with exclusive bonus features if you found out we weren't alone, if someone showed you, proved it to you, would that frighten you? Well, now the truth belongs to 7 billion people in an original event film directed by Steven Spielberg from a screenplay by David Koepp, starring Emily Blunt, Josh o', Connor, Colin Firth, Eve Hewson and Colman Domingo. Find Disclosure Day on major digital platforms now with no subscription required, also in theaters.

[Ed Zitron]

None of these businesses are profitable, not a single one of them. What's really interesting is none of them talk about the AI revenue. None of them. Microsoft mentioned it in two quarters, last quarter of 2024, first quarter of 2025, and then stopped mentioning it entirely. IBM just stopped mentioning their AI revenue. It's are they shy? And so when nobody wants to talk about the money and nobody can really precisely describe the outcomes, that's when people should get a little concerned.

[Chris Hayes]

Hello and welcome to why Is this Happening? With me, your host, Chris Hayes. Welcome back to our ongoing series about all of the implications of the AI boom. You have probably seen. Well, I don't know if you have, but maybe you've read the book the Big Short by Michael Lewis or seen the phenomenal movie by Adam McKay, which is like I think genuinely a classic. The Big Short. The tale there is a chronicle of a disparate group of what you might call our kind of financial dissidents, who in the sort of era of 2006, 2007, as the housing boom is going and everyone's making a ton of money, start to sniff out there's something very deeply amiss, deeply wrong. And the more information they get, the more convinced they become that everyone in the market more or less is wrong, that this entire money making machine is about to collapse in on itself. Because they're people in finance, they take bets that it will collapse and those bets prove to be accurate. It's a very satisfying tale and a satisfying book and movie because it's a Kind of David and Goliath, and you sort of get to follow the prophet as the prophecies come true. And the reason I bring all this up is we're in a boom right now with AI. I mean, the amount of money that's being put into it is staggering. And I think the broad amount of the financial system and the press thinks that it's going to pay off at least, you know, that's basically been what markets have priced things at. If you look at the stock of the big AI companies, and particularly Nvidia that makes the chips, this all depends on. It keeps going up and up and up. It's been a little down recently, but there are. Just like there were at the housing bubble, there are dissenters and dissidents. I'm going to talk to one today. Now, the thing about being a dissenter or dissident against this sort of conventional wisdom is you could be wrong. You know, there were people that thought, for instance, the Internet was never going to amount to anything, or that personal computers were a ridiculous technology. And in the long view of history, it's always a fine line between crank and profit. Sort of depends on what happens in the end. But given the fact that there is so much money at stake, and because I think the maximalist view that this is going to end in tears is one that people get the least amount of exposure to, I wanted to spend some time reckoning with that view today with my guest, Ed Zitron. He's the host of the Better Offline podcast and he writes the where's your Ed at? Newsletter on Ghost. And I would say that he is one of the foremost AI bears or even AI haters on the whole Internet. Is that fair?

[Ed Zitron]

I think that's fair. I really hate bad software like that. Really. I love technology. I owe my life to technology. But the current state of the tech industry sucks.

[Chris Hayes]

You're a tech person. You were a gaming journalist for a little bit.

[Ed Zitron]

Yes, I was. I wrote about video games in England, for goodness, about five, six years. Moved to America in 2008. Great year. Fantastic year to move to America. And yes, it's strange watching what's happening. And it feels like as I watch the pieces fall together, it's all been kind of working up to this. Because the AI bubble is a symptom of a larger problem with the software industry. The hypergrowth era is ending. We have seen software as this thing will always grow exponentially. So every single bubble is. Is looked at it through the same lens. You say Metaverse will grow exponentially. NFTs will take over all culture. We won't buy physical things anymore. As far as collectibles go, we'll only have these digital versions. Everything's seen through what I call the grot economy, the growth at all cost mindset. And I think we're coming to the close of it.

[Chris Hayes]

Let's stay there. Because I think one of the reasons that I hear skepticism about AI from people and I think is irrational is that we did just go through this huge hype cycle. Yeah, that was, I mean, a little bit white from memory, but before AI, during COVID there's a huge amount of money sloshing around the economy because of all of the money was being directly injected both through fiscal stimulus and the Fed. And people were spending all their times in front of their screens and there was this huge boom around the Metaverse blockchain and like non fungible tokens. NFTs.

[Ed Zitron]

Yes.

[Chris Hayes]

What do you think the takeaways from that sort of boom bust cycle are?

[Ed Zitron]

So a little bit before the Metaverse, there was another bubble that other people forgot about and that was Clubhouse. So Clubhouse was this audio only social network that if you talk to venture capitalists at the time, oh my God, this was the biggest thing ever. It was going to be worth a bazillion dollars. It was radio too. Now nobody went and looked up how much Radio makes or what the revenues were really anything but. They got celebrities on, they did everything they could. It was very clear the venture capitalists were pushing this so they could get a big acquisition. Now never happened. And Clubhouse has kind of fallen into irrelevance with NFTs, with the metaverse. People forget that Meta used to be called Facebook. We still call it Facebook. They changed their entire company name to Meta. One of the craziest things in history. And we just don't talk about it. It's insane that happened. You had people on CBS News being like, yep, the Metaverse is here. We're all going to live in the Metaverse. It's very real, it's going to happen. And the actual experience was a very bad virtual reality experience. But I feel like the tech industry has kind of been LARPing for the last 10, 15 years.

[Chris Hayes]

They LARPing, meaning live action, role playing, pretending, going through the motions.

[Ed Zitron]

Because you had the era of, of smartphones and mobile apps, huge deal. You had the era of software as a service SaaS. These were the big revenue drivers of the tech industry. It was a way to get more money out of people because you had a Subscription service, or you had an app you could buy on your phone. Most of those all have monthly subscriptions.

[Chris Hayes]

Convert to subscriptions.

[Ed Zitron]

Exactly. So it was a way of getting people away from that troublesome thing where they only paid you once. Nevertheless, this worked for a while, and then apps worked, and then nothing else really worked. We kind of started running out of kind of hypergrowth ideas. We haven't really had one since, like, I don't know, LinkedIn maybe. There are probably some examples. But we just stopped having big things that worked. So the tech industry did what it did before. Hire a lot of people, put a lot of money into things, buy a lot of things, buy little companies. The Activision Blizzard acquisition from Microsoft was claimed as a metaverse thing, which was very silly, because if the metaverse is all video games, that's just so much. But it's the tech industry doing what it thinks works, and I think we're getting to the end point of that.

[Chris Hayes]

I want to wait to talk about the tech.

[Ed Zitron]

Right.

[Chris Hayes]

And hide that off from the economics of it, but just stay on this for a second, because to me, the big difference is no one could ever really explain to me the use case of the metaverse. Right. You know, people would sometimes be like, yeah, I'd be like, well, what's.

[Disclosure Day Advertiser]

What.

[Chris Hayes]

What do I do with it? And people would say, yeah, with crypto, it's like, well, can I pay for a cup of coffee with it? No, you can't really do that. No, it's. It's essentially, there was a little bit of talk of like, the blockchain is going to replace contracts. And I was like, but is anyone actually doing that? No. And even the metaverse, you know, you kept sort of saying like, what's the use case? What's it do? No one can ever answer. I don't feel like that way with AI. Like, there are use cases. In fact, I've used it for use cases. And it seems to me like there's a much clearer, like, okay, do this doc review. Here's a thousand documents. It would take a human to go through them. It can read and synthesize. Now, the question of whether it could, like, do it well or not is a different question. But to me, the difference is that you can at least articulate what it's for in certain circumstances or reasons or places. It might be useful in a way that, for me, the metaverse never did.

[Ed Zitron]

So there are uses for large language models. If you remove all of the hype, there are things they can do. The problem is Is here's a little challenge for you. Go and talk to a bunch of AI boosters and ban them from speaking in the future tense. Don't allow them to say anything about this. Will. It might, or it could say what it does today. Because when you do that, it's not really clear what's changed. Yes, you can use it to review documents. The results are not great. Or maybe they are. You actually don't know because you didn't read the documents. A thing that is based on statistics did. The idea that you can rely on it is inherently broken because OpenAI's own research says hallucinations are a part of these things. They're never going away. You have people in the AI industry claiming hallucinations are going away. They're just wrong. OpenAI said it. You're going to argue with them.

[Chris Hayes]

There's a recent paper I was just looking at yesterday that says that even when I use AI, I do use it where you sort of gate the sources. And there's a paper I saw yesterday that even when you gate the sources, when you're saying just use these sources, you cannot purge a hallucination.

[Ed Zitron]

No. And even with coding LLMs, because this is one of the most annoying debates ever, is the usefulness of LLMs within coding. And I think the big thing with the AI LLM coding debate is you're beginning to find out that there are. I don't want to say a lot, but there is a contingent of software engineers that might not know a lot about code or might be getting by with not a ton of information. To them, this might seem magical. There's a amazing writer called Nick Suresh. He did this amazing blog called I will effing pile drive you if you mention AI again. And he made the point that this is something that has its uses for the little things. But the moment you start expanding it to building entire things for you and writing all this code for you, you are just kind of kicking the can. You're still going to have to read all this code to make sure it makes sense. Or alternatively, you could not read it and just hope it works. Software can function when the code is bad. Doesn't mean it's secure or stable or efficient or indeed that someone else coming along in the future to read it can understand the intention because there was none because the large language model wrote it.

[Chris Hayes]

Right. The sort of quality control question which you get in research seems to me a big thing. Right. But let's put that aside for a second. Right? Is a quality control question. Solvable is sort of put to the side. It seems to me that it's worthwhile to just for the next part of this to distinguish between is a tech useful or even transformational and is the current financial investment in it justified? As distinct questions, yes, and two examples come to mind. There was an enormous railroad boom that happened in the 19th century in which an enormous percentage of the country's entire GDP went into railroads. It completely overbuilt and it led to a huge crash. And that crash led to a Great Depression. It doesn't mean that railroads weren't useful. In fact, railroads are quite useful.

[Ed Zitron]

Right.

[Chris Hayes]

But it also is the case that you could have a useful technology that leads to a boom and bust. The Internet being another example. Right. It isn't the case that the Internet proved not to be useful. It's also the case that like there was a huge boom in 1999 and 2000. Right, right.

[Ed Zitron]

And I have this thing I've been saying, the beginning of history. It's not fully connected to the Fukuyama, but nevertheless it's the. I don't think it's instructive. I understand why people do it. It's how human beings work. I don't think it's useful to look at these previous booms because when you put trains on the railroad, sometimes the train didn't just randomly go up or into the ground with the original Internet systems. Back in the middle of 2025, a guy called Jim Cavallo from Goldman Sachs did a great piece along with some other analysts called Genai Too much spend for not enough gain. I paraphrase the title there. And he made the point that when the Internet did cost a lot of money at the $64,000, some Microsystem servers, nevertheless that capital outlay was completely different. But there was also a very clear path to utility. It would be the dispersion of fiber optic cable. It would be the access points, the actual things being built so that people could get to the Internet and high speed Internet on top of that. The same thing happened with smartphones. Covello notes that in the early 2000s, there were clear roadmaps, smaller Bluetooth radios, smaller GPSs, smaller chips, smaller batteries that would lead to smartphones. No such path exists for large language models. For that example to make sense, you would have to have a way in which the cost came down and the hallucinations went away. Neither of those appear to be happening. And indeed the efficacy of these models, their actual outcomes, it's actually very difficult to measure them. The Benchmarks are deliberately created for them. And all of the benchmarks for software engineering are focused on one programming language, Python, and, and very common GitHub issues. So to train for more things, they're having to create specialized data. They're going to have to do that forever. And even then it isn't obvious if it's actually fixing things.

[Chris Hayes]

Right. I mean, this is this problem of basically, are they training on the test data? Right. Are you basically saying, here, take a look at all this data and then we're going to test you on it and oh, lo and behold, your performance is good.

[Ed Zitron]

Fun fact about that. They actually found that one of the anthropic models had just started going and looking for the solution, wasn't trying to solve it, just went on GitHub and did it. Now, people mistake this for intelligence. No, you asked a thing to do a thing and it did a thing.

[Chris Hayes]

Right.

[Ed Zitron]

It's just doing the functions it was told to do.

[Chris Hayes]

Okay, but that's a great example because like a year ago it couldn't do that. I mean, it is doing something new, even if it's going to GitHub. Right. GitHub for people that don't know, is a sort of open source library and

[Ed Zitron]

where you can host projects.

[Chris Hayes]

Yes, where you host projects and people share code. But like a year ago it didn't do that.

[Ed Zitron]

Right. It used the web search tool. It used the tool it's had for a while. Perhaps it did something new, I guess. But it's a lateral improvement. It's an improvement on the thing it's already doing. It's not making unique software. Even the clawed code things. You're seeing where people are spitting out websites. There are tens of thousands of website templates and open source software projects that they're replicating.

[Disclosure Day Advertiser]

It did.

[Ed Zitron]

This is a little bit, I won't get too in the weeds here. They did something called a C compiler and anthropic said, we made this, we did this. It was a clone of an open source project and it was less efficient. It was something like 10,000 times less efficient, which is crazy. And these things don't make novel ideas because if you just look at what has already happened and say, well, based on this, this will happen. You'll never.

[Chris Hayes]

Can't get out past the arithmetic. Statistical average.

[Ed Zitron]

Exactly. Yes.

[Chris Hayes]

So let's talk about the scope of the money here. Basically paint a picture of how big this bubble is that you say is a bubble where the money's coming from and how it's flowing.

[Ed Zitron]

So it's around a trillion dollars now I think by the end of the year. If you think about all of the venture capital funding, all of the money that's been put into data centers, all of the capital expenditures from Microsoft, Amazon, Google, Meta and the money flowing through Taiwanese server companies like Hon Hai, so Foxconn and Quanta and all that, the money is coming from a few places, it's coming from venture capitalists and I can get into the crisis there soon. Private equity and specifically private credit. And actually a lot of the money is coming from Japan, Sutomo, SMBC and Mitsubishi mufg. I swear I'm going somewhere with this. But the money is coming from private equity, private credit, venture capital and in some cases the hyperscalers themselves. And most of it's flowing to like three companies.

[Chris Hayes]

I mean it seems to me that it's also. Right, so when you're, when you're talking about AI and anthropic, right, they need to raise capital. But places like Google or Microsoft are spending. I mean Google just throws off a ton of cash, right? So that's a place where they've got arguably the most profitable business in the history of human capitalism. And they, they can just sink that cash into more and more investment.

[Ed Zitron]

The problem is that's slowly not becoming true. Amazon I think is raising tens of billions of dollars of bonds. Google already did the same thing. Microsoft probably will at some point. Microsoft I think is the only one out of them that is not using debt, that's no longer just using cash flow to pay for this.

[Chris Hayes]

I see.

[Ed Zitron]

Because none of these businesses are profitable. Not a single one of them. What's really interesting is none of them talk about the AI revenue. None of them. Microsoft mentioned it in two quarters, last quarter of 2024, first quarter of 2025 and then stopped mentioning it entirely. IBM just stopped mentioning their AI revenue. It's. Are they shy? And so when nobody wants to talk about the money and nobody can really precisely describe the outcomes, that's when people should get a little concerned.

[Chris Hayes]

More of our conversation after this quick break.

[JP Morgan Advertiser]

You can't control what you can't see. JP Morgan brings clarity to your treasury operations. With near real time reports, reporting and forecasting, you get the visibility you need to help optimize your cash flow. J.P. morgan 2026 J.P. morgan Chase Co. All rights reserved J.P. morgan Chase Bank NA Member FDIC deposits held in non US branches are not FDIC insured. Non deposit products are not FDIC insured. Not all products and services are available in all geographical areas. See jpmorgan.com paymentsdisclosure.

[Ed Zitron]

You're never just one thing.

[Disclosure Day Advertiser]

You're the boss.

[Chris Hayes]

Hey Google, when's my next meeting?

[Disclosure Day Advertiser]

The athlete that class wrecks me and their mom. Everyone in the all new Mazda CX5 more to move every side of you. Learn [e-mail] Google is a trademark of Google LLC.

[Ed Zitron]

Sequences shortened and simulated.

[Disclosure Day Advertiser]

Watch Disclosure day at home now with exclusive bonus features. If you found out we weren't alone, if someone showed you, proved it to you, would that frighten you? Well, now the truth belongs to 7 billion people. In an original event film directed by Steven Spielberg, from a screenplay by David Koepp, starring Emily Blunt, Josh o', Connor, Colin Firth, Eve Hewson and Colman Domingo. Find disclosure day on major digital platforms now with no subscription required. Also in theaters.

[Chris Hayes]

If we talk, say a trillion dollars, right, and the idea is you're investing all this money. What does the investment go to? Like, what needs to be built that all this money is sunk into?

[Ed Zitron]

So there's two things to look at. There are the AI companies, so the OpenAI's and Anthropics of the world, and then the hyperscalers, software and hardware. Okay, so AI companies like Anthropic. It just came out. Krishna Rao, the chief financial officer of Anthropic, in their case against the Department of Defense, just said that anthropic, through March 2026 for its entire lifetime made exceeding $5 billion. They've spent $10 billion in that period on training and inference. Inference is the creating of the output. Fancy word for that. Training is this word that's meant to conjure up in your head this idea of research and development. Training in large language models can mean everything from pre training. So feeding a bunch of information to post training, which is everything from we're going to give you some stuff and test the outputs to minor tweaks to stop something called model drift, which is just when a model that is trained on static information will eventually become irrelevant. So you need to keep updating it to make sure when you feed it something, it understands it.

[Chris Hayes]

And there's actual, you know, huge human intervention here, which is like, no, that's wrong, that's wrong, that's wrong. Because you have to kind of train the model to learn.

[Ed Zitron]

Exactly. You've got human trainers who are training the models themselves. As in model gives an output and they go, that's a good one, that's a bad one. Then you've got people literally creating training data. Now where do they spend that money. So this is the top layer, the AI labs, those ones are spending it renting GPUs from Nvidia, which are usually in the case of Anthropic, held by Amazon or Google, or in the case of Amazon and Google, their own custom silicon TPUs for Google and Trainium and Inferentia for Amazon. Now putting all that aside, it's just spending on the chips to make the thing.

[Chris Hayes]

Right. So you've got. If I'm anthropic, I got, I got labor costs. Right. I got employees. And then I have to. To do all, all the stuff that I want to do. Run these models is very, very computation intensive.

[Ed Zitron]

Yes.

[Chris Hayes]

And in order to do that computation, I need physical hardware.

[Disclosure Day Advertiser]

Yep.

[Chris Hayes]

The so called GPUs, which are the chip that Nvidia and others make, which is this sort of frontier next generation processing chip, Right?

[Ed Zitron]

Yes.

[Chris Hayes]

And the way that it works is that Claude and AI rent that hardware.

[Ed Zitron]

So it's crazy how much it costs as well, because they want to say that inference is profitable. No one's actually proven this. It's actually quite expensive to provide a user a service. The other problem is coding models especially are incredibly computationally expensive. You've got one user who might be tying up six to 12 GPUs, each one costing 50 grand a piece or more. What's crazy and what really makes this different to most software eras is that your most excitable customers are the ones that cost you the most. And in all of the cases of the AI labs, they're subsidizing them. Claude code. Crazy fact. Researcher called Shellick found this. For every dollar that someone is spending on an anthropic subscription, when they use Claude code, they can spend anywhere from eight to thirteen and a half dollars worth of compute costs because Anthropic is subsidizing them.

[Chris Hayes]

Right. Let's stay on Claud code because this is important on the business model. Right. So Claud code is. People have been crowing about it and almost every engineer I talks about is using it. You can do things where you're basically giving it plain language instructions and it's coding for you. The back end of what it's doing is extremely compute intensive.

[Ed Zitron]

Yes.

[Chris Hayes]

And the expense of that is renting the GPUs, the electricity. Right. The server space. Right. Those are the basic.

[Ed Zitron]

Usually you pay the company like Google or Amazon directly, but that's the. Right.

[Chris Hayes]

So that's the business relationship is I'm anthropic and I'm paying some other company that's doing all that back end stuff.

[Ed Zitron]

Yes.

[Chris Hayes]

Right. And the cost of that thing I'm paying them for. Right. Can be $13 for every $1 I'm getting in revenue. I mean, yes, I think about this a lot. I use this example in another conversation just with, with Google and Gemini where if you say, what's a good Korean restaurant in Brooklyn? Google will show me a Gemini response at the top. And then there's a Reddit thread that's like great Korean in Brooklyn. Right. The computational, the like actual resource cost of just going to the Reddit thread is essentially zero, basically tiny amount. But the Gemini cost was like pretty significant to go generate all that computation in the back.

[Ed Zitron]

And you've scrolled right past that and gone straight to Reddit because you trust the person way more than you're going to trust Gemini.

[Chris Hayes]

Right, but. So the point is, even if this thing is producing use, like in the cloud code case, one of my understandings of your main argument here is that the current model is they are wildly subsidizing because the compute is so expensive and intensive, in order to make it work, they have to wildly subsidize it on the consumer end.

[Ed Zitron]

Yes. So really simple explanation. Anthropic has two, and OpenAI has this as well, two different kinds of customers. You've got a customer that pays you a monthly subscription and you pay through an API. It just means connect the model to your thingy.

[Chris Hayes]

Right.

[Ed Zitron]

Now, when you use Claude code, you're just paying a monthly subscription $20, $100 or $200 a month. And then you have arbitrary limits that Anthropic doesn't really specifically say. But if you were paying on the API, so if you're paying for the tokens directly from Anthropic, you would be paying not $200 a month, but two and a half, $2,700 a month.

[Chris Hayes]

Gotcha. So these subscriptions are essentially massively marked down to get customers who subscribe.

[Ed Zitron]

Right. But there doesn't appear to be a way that you can convince. I don't think anybody that's paying 200 bucks a month is going to go, yeah, I'll pay three grand, that sounds great. I don't think that'll happen. And what it is is an attempt to graft the previous business models and, and use the previous growth trick, which is the initiatification, the cheap monthly fee that they can then rise and then they'll find ways of undercutting you.

[Chris Hayes]

Yeah, that was going to be my next question. Right. So this Idea of you subsidize users on the front end, you sort of lose money on every customer, you get enough market share that you can then get price power and increase. This is famously what Amazon used, which, you know, lost money on every customer and every book it sold for a shockingly long period of time and achieve pricing power. And it's also Uber is another example, right, where, you know, people remember this time when Uber came about where you could take an Uber like five bucks. Yeah, I remember landing in cities when I was doing like business travel. I mean, because New York, it was

[Ed Zitron]

always like, yeah, that was muddy.

[Chris Hayes]

It was relatively expensive.

[Ed Zitron]

Yeah.

[Chris Hayes]

But still pretty cheap. But then sometimes you'd land somewhere and be like a six dollar ride from the airport to the hotel. You think to yourself, wait, how is this doesn't make any sense to me. This can't possibly be the case that anyone's making money out of this.

[Ed Zitron]

But that in comparison, it would be if like every Uber driver cost Uber $50,000 a day. The economies are just completely different. When Uber was subsidized, I think between 2019 and 2022, when they became a kind of messy, profitable, like not a great one, it was maybe 32, $33 billion, which is a lot of money. Amazon Web Services, arguably one of the most, the single most important technological innovations ever, mostly done through just money and time, though this isn't adjusted for inflation. In the 11 years from, I think 2003 onwards, they spent 38 or $39 billion in CapEx. For some context, OpenAI raised $42 billion in 2025.

[Chris Hayes]

So you're saying the scale of the subsidy here is just way bigger than those previous ones?

[Ed Zitron]

That's the point. Yes. And the underlying infrastructure, everything is more expensive and it's not getting cheaper.

[Chris Hayes]

Okay, so if we talk about the front end model makers, right, that they're subsidizing, even in this filing, anthropic 5 billion of revenue, 10 billion in expenses. Obviously that's not profitable.

[Ed Zitron]

And that's just the compute.

[Chris Hayes]

That's just the compute. So then there's the hyperscalers, which are the physical owners that are building the data centers. Right.

[Ed Zitron]

In some cases there's a lot of independent ones now that are building data centers in the hopes that AI demand arrives that doesn't exist. And people like Core Weave and Nebius and such, who would think so Neo Clouds, they just are warehouses full of GPUs that are technically data centers.

[Chris Hayes]

So let's say I'm one of those and I build A data center. Did they then have a business relationship with one of the intermediaries like Amazon, or do they directly contract with Anthropic?

[Ed Zitron]

The answer is yes. So some of them do. Some of them. It gets even more complex. We don't need to go into it. There are people that rent the data centers who then sell the stuff. But nevertheless, that's actually kind of the problem. When you look at who's paying for AI compute and you actually really go and look at who's paying the money, there are really only two kinds of customers. Anthropic and OpenAI or hyperscalers. Meta. Oh, sorry. Nvidia. Nvidia has agreed to spend in the next five years $26 billion in AI compute deals. And I don't think it's a good sign that the shovel seller is also paying for the digs.

[Chris Hayes]

No, take a second because it's going too fast. So Nvidia buying compute is weird for this reason. I just want to walk people through this real soon. They make the chip. The chip is the thing you sell to the person that's going to say, set up a data center. Right. So in the ideal world, I'm in Nvidia. I sell a chip to the data center, the data center buys it from me because I make the useful thing. And then the data center sells its compute or rents it to one of the models. Right. If you're selling the chip, why would you want to be also buying the power of the data center? And Nvidia has made a deal where they're basically going to support the construction of a lot of data centers.

[Ed Zitron]

Yep.

[Chris Hayes]

Meaning they're going to be buying their own product.

[Ed Zitron]

Essentially, yes. They're feeding money to themselves.

[Chris Hayes]

Right. So here's some money for data centers so you can buy a bunch of our chips, which is a little bit like. It seems a little bit like you're just paying yourself for something.

[Ed Zitron]

So I'll give you the one that I think is going to blow up nasty. A company called coreweave. AI Compute company. They're a public company. They lose money hand over fist and they have tens of billions of dollars of debt.

[Chris Hayes]

And they're making. What do they do?

[Ed Zitron]

They just build. They have buildings. They fill them full of GPUs. Nvidia invested in them.

[Chris Hayes]

Okay.

[Ed Zitron]

Nvidia propped up their IPO. Nvidia bought $2 billion worth of stock recently. And Nvidia is also one of their largest customers.

[Chris Hayes]

Yeah, that's.

[Ed Zitron]

Their other customers are OpenAI, Microsoft for OpenAI and Google unsurprisingly for OpenAI, I'm not kidding you. Google is renting compute from core weave rent to OpenAI.

[Chris Hayes]

Oh, wow. So there are people that are doing compute middlemen where they rent and then they rent it out to someone else.

[Ed Zitron]

And I don't want to get too deep into it be here forever. But there are also colocation companies who build data centers to rent to core weave to rent to someone else. It's really bad. When you actually look at the non hyperscaler or OpenAI compute, there's less than a billion dollars of revenue on $178.5 billion of data center credit deals done in 2023.

[Chris Hayes]

Say that again. Who has less than a billion dollars of revenue?

[Ed Zitron]

Everyone. As far as people paying to rent GPUs, when you remove all of the hyperscalers and OpenAI and Anthropic, right, it's less than a billion of revenue last year.

[Chris Hayes]

But doesn't that just mean that the big ones are driving all the business?

[Ed Zitron]

Yeah, but the big ones are also not talking about how much money they. And in fact, the big ones are losing all the money. And the ones spending the most money, Open AI is also burning so much money they need to constantly raise billions of dollars, as they just say, some of it coming from Amazon and Microsoft and Nvidia. At some point you got to wonder if it's just the same billions being cycled again and again and nobody making a profit other than Nvidia. Nvidia is just printing money, okay?

[Chris Hayes]

That's the one place. So there is one place in this that people are genuinely making a profit, which is, you know, I always use an example. I make a sandwich for $2 and I sell it to you for $4. Right? Nvidia makes a chip for X dollars and they sell it to someone for 2x or x plus y. They are definitely making a lot of money.

[Ed Zitron]

Yeah, the panini press guy, the panini press maker, yes, they are making the money, but the sandwich costs a dollar and it costs them $10 to make. It's really bad. And their only customers appear to be themselves or very small amount of AI companies, all of whom are terribly unprofitable.

[Chris Hayes]

Right? So they're definitely making a profit. And what you're identifying as the weakness is the people they're selling to are not making a profit. So Nvidia can make a. Is definitely make a profit. They're booking profits.

[Ed Zitron]

That's great gross margins as well.

[Chris Hayes]

Inarguably true. Their stock has gone up hugely because they're doing that. What you're saying is the people they're selling to are not making a profit, and at a certain point they can't keep buying if they're not making a profit. Yes, and then the people that are buying their chips that they're selling their compute power to, which are the models, are also not making a profit. And so at a certain point, the music ends and people go diving for the chairs. Because if the models aren't profitable, then they don't need the data centers. And if the data centers aren't profitable, then no one needs the chips and the whole thing collapses.

[Ed Zitron]

There's also one abstraction that makes things a little worse, which is when I say there's less than a billion dollars last year of AI compute revenue outside of the hyperscalers. What I mean by that is it doesn't suggest there's actually much revenue potential in renting an AI data center. Bloomberg report at this end of last year, $178.5 billion of data center credit deals so debt were done in America alone last year may even be more. That's a lot higher than less than a billion. The other thing is all of these data center debt deals are basically done by new companies, so all of the debt's kind of crap.

[Chris Hayes]

We'll be right back after we take this quick break.

[JP Morgan Advertiser]

What you can't see across your international payments can hurt your bottom line. JP Morgan brings clarity to your treasury operations with near real time FX reporting to help optimize your cash flow. JP JP Morgan 2026 JP Morgan Chase Co. All rights reserved. JP Morgan Chase Bank NA member FDIC deposits held in non US branches are not FDIC insured. Non deposit products are not FDIC insured. Not all products and services are available in all geographical areas. See jpmorgan.com PaymentsDisclosure so good, so good, so good.

[Disclosure Day Advertiser]

New markdowns up to 70% off are at Nordstrom Rack stores now. Stock up and save big on shoes, tops, dresses, accessories and more must haves for summer. Join the NordicLub to unlock exclusive discounts. Shop new arrivals first and more. Plus buy online and pick up at your favorite Rack store for free. Great brands, great prices. That's why you rack.

[Ed Zitron]

Ever walk past a place for rent and wish you could just take a peek inside, maybe even explore the layout. Envision the natural light streaming through the windows or plan where your vinyl record collection would go. At apartments.com you can. With tools like their 3D virtual tours, you can see the exact unit you could be living in really envision yourself in your new home with Apartments.com, the place to find a place.

[Chris Hayes]

Basically what's happening is a bunch of new entrants are saying, hey, I can find a warehouse, get a bunch of GPUs, find electricity source, make a data center. They're entering the market and they're floating debt to make these new data centers.

[Ed Zitron]

Yes.

[Chris Hayes]

With the idea that when this all takes off, they're going to have a steady diet of customers they can sell the compute to, because compute demand is going to go up and up and up. But if the demand doesn't go up,

[Ed Zitron]

then that collapses and they won't be able to pay the debt that they raised from private credit that already has issues with people not paying their debts because their due diligence wasn't so good.

[Chris Hayes]

Right. So your understanding of the sort of vector that this gets into something that's a larger financial problem, your contention, your thesis, is that there's a ton of bad debt floating around.

[Ed Zitron]

Yes. And also this is all happening in a historic downturn in venture capital and in private equity. Since 2018, venture capital has failed to on average, there are still some success stories. Of course, to have a TVPI total value put in of higher than 0.8 to 1.2 sounds complex. It just means for every dollar you invest, you get somewhere between 80 cents and $1.20 back. That's not very good. The S&P 500 have beaten the crap out of that. That's happening with venture capital. Private equity is also having the other problem, which is private equity is having trouble selling their companies. There was the massive rush in the kind of software era, the run up there, where private equity bought an absolute crap ton of software companies 30 to 40%. The co president for Polo said this recently of private equity deals done between 2018 and 2022 were for software companies, which means the private equity firm and the software company took on debt. And after that, of course, we had the 2021 era, the massive amounts of insane crazy deals, the Metaverse era, ton of really bad companies got bought for 30, 40% higher than they should have been. So you've got private equity and venture capital sitting at this time with a bunch of stuff they can't sell, which means they don't have liquidity, which means that they can't invest quite as much. And indeed they themselves might have debt they have to pay. This is happening at a time when technology and the infrastructure behind it, referring to AI, needs more money than it's ever needed. Ever.

[Chris Hayes]

So there's two parts of this I want to push on. So one is, if you think about this idea that, look, we're going to take on a lot of debt to build something out in the future that isn't profitable now but will be okay, fine. People do that all the time.

[Ed Zitron]

That's kind of the risk of investment.

[Chris Hayes]

That's the risk of investment. People do that all the time. That's the basic model here. So then the question is, okay, one is right now it's very expensive and compute intensive to do this, but maybe it won't be in the future. And what I think is interesting about that question is that might be a really good thing if that's true for Claude or OpenAI. But if that were true, it's going to be a bad thing for all the data centers and the GPUs. Right. Like the principle right now is you need a lot of computing power. The computing power is being populated with these huge physical infrastructures and enormous amounts of investment. But maybe we'll figure out a way. There's some evidence that, you know, one Chinese model has done this, that you don't need all that computer power and you can still get the same results. Even though that would seem like a great innovation at some level if that were true. It means that all of that physical infrastructure is no longer needed or valuable. Right.

[Ed Zitron]

So we can get back to the fact that it isn't getting cheaper. What Deepseek did was they trained cheaper, but the cost of inference is still going up. Because even if the model is.

[Chris Hayes]

What do you mean by the cost of inference?

[Ed Zitron]

So the cost of inference is the amount of money that it costs to create an output. So you will see that some models have got cheaper. People conflate that with the companies themselves finding a cheaper way of doing this. They've never said that. They've just brought the price down. They can afford it when they can raise 5, 10, $30 billion at a time like Anthropic just did. What Deepseat did was they were able to train a model for cheaper.

[Chris Hayes]

Right. That's the Chinese company that sort of shocked people and there was this big hit that happened to the market because of it. They were able to sort of shortcut this process.

[Ed Zitron]

Yes, because they couldn't access the latest chips. But putting all that aside, the other problem is the pre training, which, when you shove all the data in, stopped having the same results. We kind of hit the diminishing returns point. So their only way to make these models do more is was to burn more tokens. So even if a model cost comes down, you're using more tokens to do the same thing. You're spending more money as a user. We don't know what it costs them. They're all unprofitable. But to your point, you're completely right about these data centers. They also have another problem, which is takes about two years, three years to build an AI data center. Nvidia is selling new chips every year.

[Chris Hayes]

This seems like a big problem, a depreciation problem, right?

[Ed Zitron]

Well, the depreciation problem is one in the. They burn out in three to six years. We don't really know yet, but I've heard crazy failure rates like 10 to 20% within a year. But we truly don't know that. It's both the depreciation problem and the fact that. Let's take Blackwell, released kind of in 2024, but really in 2025, we still have data centers being built like Stargate Abilene out in Texas for OpenAI and Oracle that are using Blackwell GPUs that by the time that bloody thing's built, which will be 2027, they will be two to three years old.

[Chris Hayes]

Right.

[Ed Zitron]

You will have an entire data center full of obsolete GPUs. And all of the GPU data centers being built last year are going to be Blackwell.

[Chris Hayes]

Blackwell's what?

[Ed Zitron]

It's the current gen. The new gen is Vera Rubin. This is just the GPUs. So you've got all these data centers and now you've got this flood of supply of an obsolete chip. I ain't no economics now or anything, but generally when the supply increases, they have to lower the price because everyone's got it. And you're already seeing the price of renting those GPUs come down.

[Chris Hayes]

Right, because they're older chips and so they're going to the same way that like, you know, a newer car sells for more than a used car.

[Ed Zitron]

But also there are more and more of them coming online any day.

[Chris Hayes]

Right, right, right, right.

[Ed Zitron]

And also I don't even think they're profitable for the providers to run. There is compelling evidence that no one's making a profit renting them, which is crazy. It's crazy. We're all doing this and we don't know that for sure.

[Chris Hayes]

Wait, meaning folks that have the centers, the ACT centers. Right.

[Ed Zitron]

I hear, I. It's rument, so I, I can't confirm it. I heard of a data center out in North Dakota that was losing a million dollars a day that's not a good business.

[Chris Hayes]

So one problem is the timescale for building the data centers is being outpaced by the new chips. You're building things that are obsolete. There's also the threat that maybe you find more efficient ways in which you have sort of stranded assets, right? That you have all these data centers. It turns out you don't need all this compute because we've come up with a more efficient way to do it. But again, the story that the AI people are telling invest now, it's not profitable. Keep building. And if we get to something that can, for instance, do what a first year associate at a law firm does, then you have a situation. Again, I'm just. This is the case, right? The case is you've got a situation. We hire first year associates. They largely do things like doc review and they draft memos and we're going to have a model, it's going to be trained on legal stuff. It's going to be a know enterprise system that Claude charges $60,000 a year for a huge amount of revenue. Would be like the most expensive software, basically. So the business case here is you hire a first year associate for $120,000, we charge you $60,000, right? We're making a ton of money. You're saving $60,000 and it's a bummer that the first year loss or gets out of a job. But if we could do that at scale, if there's millions and millions of these kinds of jobs that people are making high five figures to six figures that we can sell you software to replace. I mean again, this is the contention of why it would be valuable. This is the core contention. Like if you look into what these companies are saying, I guess the question then becomes is that a plausible outcome? Because I think if it is plausible, you could probably make the math work. And if it's not plausible, then you can't.

[Ed Zitron]

So the law firm example is great. The problem is I don't think enough people know what people do at jobs. Law firm associates make law firms work. Law firm associates are doing the work that partners don't want to do. And if any partners are listening, you know I'm bloody right. So what you're describing there would be AGI that just this conscious computer, which by the way, everyone's real excited to control a conscious creature. That's just describing slavery. It's important to say what AGI is. It is describing slavery, right?

[Chris Hayes]

You're saying if you achieve what they call artificial general intelligence. We actually just had a Conversation about this with David Chalmers about consciousness, that then you're actually. There's all sorts of moral implications of what that device is once you.

[Ed Zitron]

It's a slave, right?

[Chris Hayes]

Yes.

[Ed Zitron]

Back to the law slave. So this theoretical thing. Yeah. If it could do literally what a associate did, sure. But an associate does much more than just doc review. They're doing a bunch of research. And it's not just. I found a thing. Right. Look, it's drafting motions. If you get a motion wrong, a judge will sanction you and you will embarrass the partner. What you're paying for with employees in many cases is actually risk management. Judgment and judgment and taste and culture and also risk management. You are handing the risk off to a human being that you can rely on and train. Also, how are we going to make partners if we can't make associates? We're just going to hire a law student to become a partner. I mean, I don't know. I could sit around handing out people's work and talking. Now, that's. That's not true. Partners do all sorts of work. I'm sure. But nevertheless, yeah, in theory, if you could replace 10, 60,000, $150,000 in the case of a law student, you can replace 10 of them with $60,000. Sure. It isn't doing that. And large language models are sold as the reason I mentioned the thing earlier with AI boosters. They need to be legally banned from saying in the future it will could.

[Chris Hayes]

Right.

[Ed Zitron]

We need to talk about what's happening today. It isn't doing it. It isn't doing it. And in fact, every single example I hear of in specifically law large language models being used ends up with someone getting in trouble with a judge. I think they just had a DOJ person that this happened to as well.

[Chris Hayes]

Well, I don't think that's true. That every example. Because the people are using AI all over the legal world, I can tell you. But there definitely have been hallucinated citations that have been filed. And I think in some cases even by government lawyers, the DOJ that have been caught that where they're citing to a case that literally was invented by the AI.

[Ed Zitron]

So to your point.

[Disclosure Day Advertiser]

Yeah.

[Ed Zitron]

If you could do the thing it doesn't do and has no proof of doing. Yeah, sure. Grandmother had wheels, she'd be a bicycle, and so on and so forth. A lot of this, in fact all of this is really sold on the coulds and shoulds and wills.

[Chris Hayes]

Yeah. It's a bet about what its future

[Ed Zitron]

capabilities are based on what I would describe as semiotic logic even. It's this idea that because things have worked this way in the past, it'll happen before. There was a time when the Internet was slower than everyone's Internet connectivity went up. Not really the same thing because the technology was always there to get it fast. The fiber optic cable was there.

[Chris Hayes]

Right.

[Ed Zitron]

The massive overbuild was there. This is not a problem that you solve by having more compute. It is not.

[Chris Hayes]

I mean they think it is. Right? I mean, just to be clear about what the disagreement is, their contention is that they have found a reliable and straightforward law of scaling, which is that the more compute you have, the better it gets and the more that it starts to act in ways that are

[Ed Zitron]

intelligent except the scaling laws are broken. That diminishing returns. I mentioned earlier, it's no longer getting the same kind of improvements just by pre training them.

[Chris Hayes]

But isn't it? I mean, I just got to say like this is where my experience of use of the models is that they're getting much better at what specifically? Multi step tasks in research. Here's a great example. You go to clotting, you say, I said this the other day. I'm trying to figure out the relative homicide rates in major American Cities in the 1890s. I want to look at New Orleans, which is what I'm writing about, and compare it to New York and Philadelphia a year ago. With previous models you would have gotten essentially nonsense or you would have gotten like, well, here's the Wikipedia, here's a few things. In this case it like went through, it found like there's two like real sources on this. Like there's a book about southern homicides. There's another book about northeastern policing. I know this because I've actually done the research. Right, right. It goes through. It basically does find in one of the books because it's in public domain what the New Orleans homicide rate is. It talks about what the data difficulties are in New York, Philadelphia and it basically spits out an answer that I can check because it's citing it. That's basically correct.

[Ed Zitron]

Okay.

[Chris Hayes]

In New Orleans it's 25 out of 100,000 and in New York and Philadelphia it's 5. Something like that. A computer could not do that a year ago. Like it just couldn't. Now there's all sorts of ways in which I can check it because I have the expertise. But this was like a sophisticated multi step thing that I had to go through and sort of use a bunch of powers that it just didn't have. A year ago.

[Ed Zitron]

I mean, you had to check every step though, didn't you? You had to go and check all the data.

[Chris Hayes]

I did have to check the citations, yeah.

[Ed Zitron]

So what you're describing there is an improvement. They have found ways to connect them to web search tools.

[Chris Hayes]

Right.

[Ed Zitron]

These things are able to drag stuff. But what you're ultimately describing is more sophisticated but less reliable search. It is an improvement because they're able to post, train it in that case and say, this result is bad, this result's good. I've used even the most sophisticated ones used by hedge funds, the searches. The problem is is that, yeah, it will get some things right and it will find the occasional thing, oh, you didn't see this in a 10k from 2 years ago. Problem is you have to check every single bloody thing. You can't rely on anything. Perhaps it helped you get in the right direction. Is that worth this much money? Is. I mean what you're describing you were seeing in models middle of 2025, I guess we did something.

[Disclosure Day Advertiser]

Wow.

[Ed Zitron]

We have better search, more sophisticated but

[Chris Hayes]

less reliable search well or multi step things. I mean that the thing to me was that this is a fairly compound task, right? So it has to do, it's got to do a bunch of stuff. And the thing that I thought was striking was that it actually, it did a good job of finding the right source, which that was sort of interesting to me. Like, oh, that is the book, you know, that is the book where this is contained. You didn't just like go to the Wikipedia page. I think the thing that I kind of come back to, and this is the sort of horns of the dilemma, and many people have talked about this, is that it seems to me that there's no way out of some kind of cataclysm for this reason.

[Ed Zitron]

How do you mean?

[Chris Hayes]

Either your case is correct, in which case it's just not going to be profitable and the whole thing is going to collapse in on itself.

[Ed Zitron]

Yeah.

[Chris Hayes]

Or you're wrong and they get a lot better and they are profitable and what being profitable means is that they can replace the labor of tens of millions of people.

[Ed Zitron]

Right.

[Chris Hayes]

That's another cataclysm. Like the point is that like if they're right, if the thing that they're promising, which is like, oh yeah, we could start getting rid of all these people that do all these jobs and replacing it with AI and that's good for the profitability of these companies, but I think it's probably insanely destructive to the macro economy in American society.

[Ed Zitron]

Sure. And I'm not afraid of that because

[Chris Hayes]

you just don't think that's gonna happen.

[Ed Zitron]

I see no signs of it.

[Chris Hayes]

Right. But it is the only way.

[Ed Zitron]

Like, it is the only way it would make sense.

[Chris Hayes]

That's the point. The point is that, like, for the math to work out, it has to be something pretty darn revolutionary and it

[Ed Zitron]

has to be trillions of dollars. Like, I worked it out mathematically by 2030, for any of this to make sense for Microsoft, Meta, Google and Amazon, they need $2 trillion of new revenue. Not enhanced revenue. I mean brand new, brand spec and new dollars in a software industry that has never been higher than $700 billion of yearly revenue.

[Chris Hayes]

And what is annual U.S. gDP is like $30 trillion. Right. So you're talking about like just enormous.

[Ed Zitron]

Just 10% of the world GDP.

[Chris Hayes]

Right. An enormous part of the entire economy.

[Ed Zitron]

And it needs to happen pretty quickly in the next six months. There is one other thing, though.

[Chris Hayes]

Yeah.

[Ed Zitron]

I think that there is a social contagion that will happen with this. Look around the world of bosses right now and the amount of them who are like, oh, yeah, I can't wait to replace everyone. I want to replace all the actors in my movies and replace all my workers. He's just going to give me money and then I'm going to have all the money and the pieces of crap I sell things to. There's hogs braying for my slop, the excitement in it. But also, how many of them are just wrong? How many of them just say things that aren't true? We have people in newspapers saying things about A.I. that aren't true. We have bosses claiming things about A.I. that aren't true. We have people lying about it. It's truly obscene. And regular people know regular people. Like, if you go and talk to, like, electricians, H Vac people, hairdressers, teachers, their reaction to this is horror. There are some who are using it to cut corners. Everyone wants to do that. Human beings do that.

[Chris Hayes]

Well, there's also a lot of people that have, like, crazily intense parasocial relationships and talk to it all the time.

[Ed Zitron]

I think that there should be criminal tribunals for the companies that it's disgusting anyway. I think that we are going to see something happen before the economic stuff as an outcome of it, actually, where regular people have seen how their bosses think of them. And it's happened for years. You saw it with remote work where bosses were like, hey, you got to get back to the office, man. You got to get back there. I got to be able to look at you every day. I got to be able to stomp around so you can feed off my mood. You got to go to the metaverse now, because that's where I'm going to be. Have fun staying poor. You weren't in crypto. Also, I'm replacing you with AI. So you've got that and then you've got the other thing, which is it needs to make all this money now. Now, next six months, OpenAI even then raised $110 billion. Actually, they only raised 15 billion. 35 billion of the money from Amazon is due when they achieve AGI or go public. And both the 30 billion from Nvidia and the 30 billion from SoftBank are being paid in $10 billion tranches like Klarna. Not literally, though. And what's funny is SoftBank has to raise $40 billion in loans to pay for their part. Everything that's happening is a stress test of debt and equity. How much can venture capital spend? How much can hyperscalers afford? How much money is left in the coffers?

[Chris Hayes]

So then what is that out of your theory? What emerges as a prediction of the first place that you'll see a crack? A fault? Like what. What would be the first sign?

[Ed Zitron]

We're already seeing it with private credit, so I kind of hinted at it earlier. Private credit, private equity massively bought so many different software companies. And when they bought them with these leveraged buyouts, they bought them, pumped them full of debt, and then took on debt to buy them. I've heard something ridiculous like private equity firms are leveraged to four to six times the value of their assets. So you're already seeing it. There was a stat that came out the other day. There's $42 billion of software loans just for software companies that are in distress status. So not likely to be paid.

[Chris Hayes]

Right.

[Ed Zitron]

You have, across the board, you can go and look inside that. They have to publish this. Private equity firms, private credit firms, and BDCs, business development companies, basically the same thing. They're suddenly starting to take payment in kind for loans. As in you get stock, you get given stock, and then you just kind of put all the cost onto the end of the loan. They're not getting paid on these loans. These loans are going to. They're starting to default.

[Chris Hayes]

So you think you're going to start to see loan defaults, like the private credit market is going to ripple for already seeing it.

[Ed Zitron]

And then I think my real, my three horsemen are you're going to see a Datacenter project fall apart before it's complete. You're going to see in construction one collapse and then you're going to see a fully constructed one that has to shut down because it runs out of money. Because remember, these things are heavily debt billed. They are full of debt. There's not a single one of them that is even close to profitable before the debt. And then you add the debt on top.

[Chris Hayes]

That's interesting. So those three things look for that in data centers.

[Ed Zitron]

Exactly. Because the AI bubble, in my opinion, is a symptom of the larger death of software as a growth model. Because the assumption was software eating the world. Mark Andreessen was that every industry could be software ties, which is true, and that as a result all of them could grow forever. Private equity venture capital bought into this. They invested in all these companies. Except now nobody will buy the companies. M and A has died and it's harder to take them public. And if you can't take it public and you can't sell someone else, what do you do? Well, the answer is you sell them to another private equity firm. So there's currently a game of hot potato called continuation funds or secondaries, depending on who you ask. Eventually no one's going to have the money to buy the thingy.

[Chris Hayes]

The other thing is that at the core of all this, whatever financialization you're doing, you have to make things that are profitable at the bottom of it.

[Ed Zitron]

Yes, you do.

[Chris Hayes]

I mean, that's whatever financialization happens, you know, you can have periods of where you buy an asset and you bloat it with debt and you sell it to someone else, right? And there are ways, in some intermediary sense that people can make money off passing things hither and yon, but in

[Ed Zitron]

the end you got to make money.

[Chris Hayes]

Things have to be useful and make money for everything to keep going.

[Ed Zitron]

And the thing is, all of these investments, and Apollo's John Zito, I think his name is, he said that the whole thing is, is that these software companies were acquired or invested in based on the idea that they would grow like they did between 2005 and 2018. Now they grow like after 2018 there's less money and there's only so many software companies you can build, only so many people you can sell to.

[Chris Hayes]

The last thing to me that also seems possible just to end it here, is that, you know, there's this sense that like sometimes they'll talk about their own vision of artificial intelligence being like a utility, like electricity or water. And what's interesting about that is that, like, utility companies aren't that profitable and they're not that sexy and they're not like. It's a strange thing because it seems to me like maybe it is possible this becomes like a utility that, like, everyone sort of has it or has access to it. But if. If that's the case, then it's basically just a commodity. Like, utility companies are not super profitable. In fact, they're regulated.

[Ed Zitron]

And that's the thing, the utility thing doesn't really make sense because water is water. An AI model is many different things run by many different companies that needs constant maintenance to avoid model drift, because otherwise it's just a static object that cannot respond to new things. You don't need to continually make sure power is power before it turns into something else. Yes, there are power of regulation things, I know, but one other very scary thing to add, I don't like scaring people, but some of the money that's going into data centers now, more and more, in fact, from like, Blackstone and so on, and Aries, I hear, is from insurance funds and retirement funds because they've needed more liquidity than they had. So they've started investing in private credit loans to data centers. And the sales pitch is simple. This is the future. This is a good yield. You'll get paid more on your money without the worrisome thing of will they always be able to pay back their debt? And we mentioned Oracle earlier. Oracle is a mess, but Oracle has taken over $100 billion of debt. They had negative cash flow of $24 billion, and they are building 4.5 gigawatts of data centers, so hundreds of billions of dollars. And they're building them for one company, OpenAI, who lost at least 8 billion, more like 10 or $15 billion last year and expects to burn $230 billion by 2030. Oracle will die if OpenAI dies. And this is not me catastrophizing, mathematically speaking, Oracle cannot pay the debt if OpenAI does not make more money than Nvidia does right now by 2030.

[Chris Hayes]

Yes, that's the bet. And they'll say it themselves. Dario would say this and Sam Alt Dario made is that their bet is on a revenue trajectory that is essentially unprecedented in human capitalism. They think they can achieve that, and if they don't, it doesn't work. Ed Zitron, host of the Better Offline Podcast, writes the where's your Ed At? Newsletter on Ghost. Great to have you here.

[Ed Zitron]

Thank you for having me.

[Disclosure Day Advertiser]

Foreign,

[Chris Hayes]

You can get in touch with us by emailing withpod Gmail.com why is this happening? It's produced by Donnie Holloway and Brendan Amelia. It's engineered by Hazik Bin Ahmed Farid, Bob Mallory and Mark Yoshizumi. Katie Lau is our senior manager for audio production. Our coordinating producer is Franny Kelly. Aisha Turner is executive producer of msnow Audio. New episodes come out every every Tuesday. You can Watch us on YouTube by going to ms.dot now withpod.

[JP Morgan Advertiser]

Managing your working capital with fragmented platforms makes it hard to optimize your cash flow. JP Morgan's Working Capital Accelerator brings payables and receivables financing together into one centralized platform to help unlock your working capital. J.P. morgan 2026 J.P. morgan Chase Co. All rights reserved. J.P. morgan Chase Bank NA Member FDIC deposits held in non US branches are not FDIC insured. Non deposit products are not FDIC insured. Not all products and services are available in all geographical areas. See jpmorgan.com paymentsdisclosure.