AI Skeptic: This Business Makes No Sense

Summary

Prof G Markets — “AI Skeptic: This Business Makes No Sense”
Episode Date: May 14, 2026
Hosts: Ed Elson (for Scott Galloway), with guest Ed Zitron
Podcast Network: Vox Media

---

Episode Overview

This episode dives deep into the ongoing AI investment boom, examining whether the sky-high valuations and extraordinary spending by big tech on artificial intelligence are justified—or whether the business model is fundamentally flawed. Host Ed Elson is joined by Ed Zitron, a prominent AI skeptic and author of the “Where’s Your Ed At?” newsletter, for a candid and data-backed analysis of AI economics, industry circularity, and echoes (or lack thereof) of the dot-com bubble. The conversation also briefly touches on the geopolitical scene, with a segment on President Trump’s high-stakes summit with China's Xi Jinping.

---

Major Discussion Points & Insights

1. AI Economics: The Revenue Mirage (04:38–11:47)

• AI Profitability & Revenue Transparency:
  Zitron argues that, across the board, public companies are not showing real profits or transparent revenue streams from AI investments.

  • “Across the world, across the board with public stocks, nobody is showing a profit from AI. And for the most part, nobody is showing even the revenue they’re making from AI.” — Ed Zitron [04:38]
  • Much of the revenue reported (e.g., Microsoft’s $37B AI run rate) is money cycling between the same major players (OpenAI, Anthropic, Microsoft, Google, Amazon).

• Circular Revenue and “Fake” Growth:
  Zitron deep-dives into how AI labs like OpenAI and Anthropic rely on advanced sales of compute capacity, funded by venture capital, not sustainable customer demand.

  • “The whole thing is very circular... outside of OpenAI and Anthropic, there’s not really any sign that there's a real revenue stream here.” — Ed Zitron [05:35]
  • He questions Anthropic’s self-reported revenue numbers, citing inconsistencies and possible inflation from advance token sales: “I think both OpenAI and Anthropic are inflating their numbers.” [09:41]

• Infrastructure Shortfalls:
  The data center build-out isn’t keeping up with CAPEX claims; actual build rates are low and create hard limits.

  • “I have on good authority that one of the major hyperscalers only has around 1.5 gigawatts of actual IT capacity despite hundreds of billions of dollars of capex.” — Ed Zitron [06:01]

---

2. Skepticism on Long-Term AI Industry Viability (11:47–19:30)

• Comparisons to Dot-Com Bubble:
  Zitron contends that, unlike the dot-com era, the “residue” of AI investment (hardware, data centers) will have little after-market value if AI fails to deliver profitable models.

  • “Anthropic and OpenAI also do not have assets. They don’t have capex... They have their models, they have their model weights... they don’t really have... there’s not really a thing to recover.” — Ed Zitron [18:18]
  • The AI startups (“wrappers”) are raising and burning cash, but “not creating any useful residue.” [17:51]

• No Evidence of Broad Demand:
  Only two real “winners” (OpenAI and Anthropic), and even they are largely burning through VC cash.

  • “Other than Anthropic and OpenAI, why are there no other AI winners? Why is everyone piddling around 100 to 300 million ARR… for what is meant to be an industry changing thingamajig?” — Ed Zitron [10:41]

• Unprofitable Core Business:
  Most actual user costs are massively subsidized: “I think basically every AI startup is unprofitable at its core and there is nothing that’s going to shift these favor.” [11:29]

• Nvidia is the Only Clear Winner:
  All roads (and profits) lead to Nvidia: “They're all feeding money from Jensen Huang to Jensen Huang." — Ed Zitron [16:32]

---

3. Is AI a New Utility or a Flawed Product? (19:30–27:19)

• Utility Model vs. Traditional Tech:
  Elson wonders if AI will become an essential utility, though perhaps only sustainable with monopoly dynamics and government-like regulation.

• Dot-com Parallels Don’t Apply:
  Unlike with fiber or servers in the dot-com era, AI’s hardware investments have smaller potential for re-use, and higher ongoing operational costs.

  • “If data centers end up not being built, it's going to cost just as much money to finish them in 10 years as it will today.” — Ed Zitron [13:35]
  • “This is the big argument… people are saying we’re in a bubble now. But don’t worry, after the dot com bubble people picked up stuff for cheap. What stuff?” — Ed Zitron [19:34]

---

4. Use Cases, Subsidies, and the Real Value of AI (21:41–27:19)

• AI’s Value is Real—But Subsidized:
  Zitron acknowledges AI is useful for coding, but says most use is artificially cheap—especially via products like GitHub Copilot.

  • “There is value in coding. I’ve said that for years… but all of that use and that utility is subsidized.” — Ed Zitron [21:45]
  • "I'd love to see how popular this was if people actually paid what it cost. Right now this all-you-can-eat nonsense is inflating the value of it.” [22:55]

• Unsustainable Customer Economics:
  Companies blow through token budgets quickly, signaling high costs per use:

  • “ServiceNow just blew through their AI yearly token budget in what, five months. Uber burned through it in four.” — Ed Zitron [24:11]
  • “There is no sign that inference is becoming cheaper. There is no sign that anyone has any plan to do so.” [11:29]

• Analogy to Uber/Venture Subsidies:
  Elson compares AI to Uber's early, VC-fueled days. Zitron refutes this, saying AI subsidization is quantitatively worse:

  • “This would be like if Uber paid for the cars, paid for the gas. The gas was giraffe blood.” — Ed Zitron [26:34]
  • “You’re paying $20 to drive 100 miles a month. Except in reality, each mile is costing Uber 25 bucks or more.” [27:12]

---

5. Closing Thoughts: Uncertain Future for AI Profitability (25:46–28:07)

• Scaling Won’t Necessarily Create Profit:
  VC thinking is that scale will make the economics work. But Zitron is unconvinced, especially as compute and energy costs rise, and “training” remains an endless, expensive necessity.

  • “There is no profitability for any of this and you have to keep training the models because of model drift.” — Ed Zitron [25:46]

• Elson generally agrees with the skepticism about pace, path, and the difference from past technology cycles, but is less sure about writing off the industry entirely.

---

Notable Quotes & Memorable Moments

• On the Circularity of AI Revenue:
  “Much of the revenue comes from OpenAI, feeding it money from, well, OpenAI’s venture capital funders and probably leftover Azure tokens.” — Ed Zitron [04:38]
• On Hyperscaler Hype:
  “I have on good authority that one of the major hyperscalers only has around 1 1/2 gigawatts of actual IT capacity despite hundreds of billions of dollars of capex.” — Ed Zitron [06:01]
• On Subsidized User Value:
  “They were selling $15 for a dollar... This is direct money to Anthropic and OpenAI. This is not something... oh, they can claim they're profitable on inference. This is just straight up token boom.” — Ed Zitron [22:15]
• On the Real vs. Imagined Cost of AI:
  “If this was being charged at cost... I don't think people would think it was quite as useful.” — Ed Zitron [23:05]

---

Key Timestamps for Major Segments

• [01:57] – Market update and setting up the AI conversation
• [04:38]–[11:47] – Zitron challenges the AI revenue/profit story
• [11:47]–[19:30] – Dot-com versus AI bubble, asset and residual value debate
• [19:30]–[27:19] – AI as utility, subsidized economics, the Uber parallel
• [27:19]–[28:07] – Closing thoughts on AI’s future

---

Bonus: US–China Geopolitics Segment

• [32:03] Onward: Analysis of President Trump’s summit in China (with expert James King)—focus on trade, Taiwan, rare earth minerals, and the current economic leverage China holds over the US due to recent events in Iran and global supply chains.
  • “If you ask me, this is the first summit in US-China history where the Chinese president has the upper hand and Trump is going there asking, rather than the other way around.” — James King [38:58]

---

Episode Takeaway

This episode delivers a sharp, skeptical, and data-driven challenge to the AI investment gold rush. Ed Zitron’s view is clear: despite record-breaking spending and revenue claims, the fundamentals don’t add up, the revenue is often circular or inflated, and both the market and the public may be in for a rude awakening. The conversation is engaging, punchy, and essential listening for anyone thinking critically about what drives the current AI boom—and whether it can ever deliver on its financial promises.

---

Transcript

[Adobe Announcer]

Adobe Acrobat your new foundation. Use PDF spaces to generate a presentation. Grab your docs, your permits, your moves, AI levels of your pitch gets it in a groove. Choose a template with your timeless cool. Come on now, let's flex those tools drive, design, deliver, make it sing. AI builds the deck so you can build that thing do that, do that, do that with Acrobat learn [e-mail] do that with Acrobat

[Windows 11 Advertiser]

Study and play come together on a Windows 11 PC and for a limited time, college students get

[Ed Zitron]

the best of both worlds.

[Windows 11 Advertiser]

Get the unreal college deal everything you need to study and play with select Windows 11 PCs. Eligible students get a year of Microsoft 365 Premium and a year of Xbox game. Pass ultimate with a custom color Xbox wireless controller. Learn [e-mail] studentoffer while supplies last ends June 30 [e-mail] when you need to

[Indeed Advertiser]

build up your team to handle the growing chaos at work, use Indeed Sponsored Jobs. It gives your job posts the boost it needs to be seen and helps reach people with the right skills, certifications and more. Spend less time searching and more time actually interviewing candidates who check all your boxes. Listeners of this show will get a $75 sponsored job credit@ Indeed.com podcast. That's Indeed.com podcast. Terms and conditions apply. Need a hiring hero? This is a job for Indeed Sponsored Jobs.

[Ed Elson]

Today's number two. That's the number of times Sir David Attenborough has been knighted for his services to British television and conservation. It appears that after Brexit, the UK has run out of people. Tonight,

[AI Industry Analyst]

money markets matter. If money is evil, then that building is hell.

[Adobe Announcer]

The show goes on. Sell sell.

[Ed Elson]

Welcome to ProfGen. I'm Ed Elson. It is May 14th. Let's check in on yesterday's market vitals. The major indices were mixed as investors digested new inflation data from the producer price index. Tech stocks drove the S and P and the NASDAQ to new records while the Dow fell. We'll talk about that inflation in a moment. Inflation concerns also pushed the yield on 10 year treasuries higher on Kalshi. The odds of a rate hike this year have steadily climbed to 31%. Meanwhile, Apple stock closed at a record high as CEO Tim Cook joined President Trump as well as several other big tech leaders for a summit in China. We will be discussing that shortly as well. Okay, what else is happening? Big Tech is on Track to spend $725 billion on AI in 2026, and it shows no signs of slowing Yesterday, reports surfaced that Anthropic is in talks to raise $30 billion at a valuation north of $900 billion. And SoftBank posted a $46 billion gain for the year, fueled by the soaring value of its stake in OpenAI. The markets are currently rewarding the AI boom. The Nasdaq is up 27% since its March 30 low. Much of the gain comes from the AI trade, with chip stocks recording their best monthly performance in decades. But not everyone is buying it. Big tech's free cash flow has been shrinking as AI capex balloons. And some investors are when, if ever, the spending will generate a return. So we wanted to check in with one of AI's fiercest skeptics and critics and ask what could the market be missing? So, joining us to talk about the state of AI and perhaps some of the stuff that people are not paying enough attention to, we are speaking with Ed Zittran, author of the where's your Ed at? Newsletter. Great name and the better Offline podcast. Ed, thank you for joining me on the show. Our producer Claire mentioned this to you offline, but I will say it now. You are probably the most requested guest we've ever had on the show. My audience has been asking over and over again, you have to talk to Ed Zitrin. You have to talk to Ed Zitron. We're finally here, so thank you for joining me. So I want to start with maybe just a summary of your views. You have been writing for a long time that you believe that a lot of what we're seeing, AI, is fake, misleading, not what we think it is. If you could just start with a summary of what you believe is really going on in the AI world right now.

[Ed Zitron]

So across the world, across the board with public stocks, nobody is showing a profit from AI. And for the most part, nobody is showing even the revenue they're making from AI. When I say revenue, I do not mean profit. Microsoft revealed that a 37 billion doll ARR run rate for AI. The majority of that comes from OpenAI, feeding it money from, well, OpenAI's venture capital funders and probably leftover Azure tokens from the 30 billion, $13 billion in funding. Anthropic similarly is funded entirely with venture capital, and their connected counterparties are seeing massive boosts in their remaining performance obligations. So their revenue backlog, mostly From Anthropic and OpenAI, in fact, between Microsoft, Google

[AI Industry Analyst]

and Amazon, $748 billion of their upcoming

[Ed Zitron]

revenue, over half of it comes from two companies either buying compute or In

[AI Industry Analyst]

Google's case, anthropic buying TPUs from Google

[Ed Zitron]

to rent back from Google.

[AI Industry Analyst]

The whole thing is very circular. And really outside of OpenAI and anthropic, there's not really any sign that there's a real revenue stream here. I have serious questions about the way

[Ed Zitron]

they count revenue, but as private companies, it's hard to pierce.

[AI Industry Analyst]

There's also the other real problem Big tech is between Meta, Google, Amazon and Microsoft. Over $800 billion of CapEx. I don't think data centers are being built at anywhere near the rates that people think.

[Ed Zitron]

I have on good authority that one of the major hyperscalers only has around 1 1/2 gigawatts of actual IT capacity despite hundreds of billions of dollars of capex.

[AI Industry Analyst]

And data centers are just not getting built. 18 to 24 months minimum. People see the AI boom as something that's happened, as in something that's there's

[Ed Zitron]

tons of data centers being built. I've seen people saying tens of gigawatts built every year. It's nonsense. I think there may only be a gigawatt or two built every year.

[AI Industry Analyst]

The compute constraints people are facing are not a result of incredible demand, but

[Ed Zitron]

a lack of actual data center capacity coming online.

[AI Industry Analyst]

We are yet to get to the

[Ed Zitron]

real monstrosity that this has created and when it pops, it's going to be very bad for everyone involved.

[Ed Elson]

So I'm going to play a little bit of devil's advocate. I want to couch this by saying a lot of what you're saying I agree with. I agree that the revenue is circular in a lot of ways. I agree with you. Your points about the data centers, the fact that they're not coming online. And there's a lot of questions on are these data centers even going to get built? Are they even going to work? Are we even going to allow them to get built through regulation and through policy? I agree with a lot of that. At the same time though, I do look at the growth of some of the revenues of some of these AI labs. I look at OpenAI and to your point, we haven't gotten the financial audit of these companies. These are privately reported, but that's the way private companies work. But OpenAI was $2 billion in ARR in 2023. It's up to $25 billion in ARR today. Anthropics reached $30 billion ARR. It took Salesforce about that. They've done it in under three years. I mean, that's. Those are dollars that are coming in and companies are trying to adopt AI as much as possible and they're paying money to do it. And I guess the way that I feel about it is we're probably very early to this game. They're doing as what, whatever they can to get this through. They want to adopt AI, But I still find those numbers striking. And to me it tells me that something about this is real and meaningful. Maybe there's some BS around around as well, but overall, I don't want to write it off. What would you make of that?

[Ed Zitron]

So let's start with that statement about being early.

[AI Industry Analyst]

We're not early. If we do it by time.

[Ed Zitron]

It's been four years been. Yeah. Coming up on four years since Chat GPT came out.

[AI Industry Analyst]

And if you think about what constitutes

[Ed Zitron]

early early, it really is investment and innovation.

[AI Industry Analyst]

We have had all the King sources,

[Ed Zitron]

all the King's men trying to make generative AI into something that generates profit. Hasn't happened yet. No one has happened it to.

[AI Industry Analyst]

I also have serious questions about how

[Ed Zitron]

Anthropic is doing revenue.

[AI Industry Analyst]

Krishna Rao, their CFO in the Department

[Ed Zitron]

of War lawsuit, actually said in an affidavit on March 9, 2026 that they

[AI Industry Analyst]

had made $5 billion in lifetime revenue. That does not match up with any of the reported previous annual recurring revenue

[Ed Zitron]

or ARR or run rate or whatever. They use numbers, it just doesn't.

[AI Industry Analyst]

When you add those all together, it's six or seven billion dollars, maybe more. I think Anthropic is counting revenue in a way that is different, fundamentally different

[Ed Zitron]

to how most companies count it.

[AI Industry Analyst]

I My theory is that when so

[Ed Zitron]

with the largest clients they have clients don't pay in arrears. Smaller ones do, but the largest don't, they buy massive amounts of tokens in advance. I think Anthropic, for example, if someone is going to do $50 million worth of tokens and it lasts over six months, they're going to take that money up front. They're going to say wow in that

[AI Industry Analyst]

month because all ARR is is month

[Ed Zitron]

times 12 or 13, depending on how they do it. So I think both OpenAI and Anthropic are inflating their numbers.

[AI Industry Analyst]

I actually think one or both of

[Ed Zitron]

them is misleading investors, but that's something we'll find out about in the future.

[AI Industry Analyst]

I also fundamentally just don't see evidence outside of these two companies that anything's happening. The largest company within this space was

[Ed Zitron]

Cursor and now they've been kind of sort of not really bought by Elon Musk.

[AI Industry Analyst]

Also, if there was so much demand. By the way, why did Elon Musk give up an entire data center to Anthropic? There are enough signals here that suggest there is a fundamental weakness and no real major business model here. We still don't know Anthropic's true burn rate. But we do know that if all of the commitments from Amazon and Google

[Ed Zitron]

come together on this $50 billion round, it's 30, it's 50, it changes every day.

[AI Industry Analyst]

It seems they'll have raised $108 billion in the last year. What's going on? Where is that money going and to whom is it going to other than

[Ed Zitron]

Google, Microsoft and Amazon?

[AI Industry Analyst]

So one of my simplest points as well is other than anthropic and OpenAI, why are there no other AI winners? Why is everyone piddling around 100 to 300 million ARR which is 10, 20, 30 million dollars a month? It's not actually that much for what is meant to be an industry changing thingamajig. And there are signs that the economics don't work. The biggest being that Microsoft GitHub co

[Ed Zitron]

pilot by the way, one of the

[AI Industry Analyst]

largest clients of Anthropic is moving to

[Ed Zitron]

token based billing on 06-01-2026.

[AI Industry Analyst]

They have been subsidizing tokens for their users to the tune of. I saw one person who spent $5,000 worth of API calls on a $39 a month program. I think this problem is across the industry. I think basically every AI startup is

[Ed Zitron]

unprofitable at its core and there is nothing that's going to shift these favor. There is no sign that inference is becoming cheaper. There is no sign that anyone has any plan to do so.

[AI Industry Analyst]

And neither Anthropic nor OpenAI seem particularly concerned with bringing those costs down.

[Ed Elson]

So some really interesting points in there. I would still disagree with your point that it isn't early. I mean when we're talking about technologies of this size, I still think four years is early and to the point about profitability. Like you know, you look back to the Internet, it took a long time for the Internet to get running. Amazon took nine years before it was consistently profitable. And eventually, I mean it, you can take decades before your business model model works properly and we have seen that with companies in the past.

[Ed Zitron]

Which ones specifically?

[AI Industry Analyst]

Just because Amazon Web Services was profitable

[Ed Zitron]

in nine years, the total capex between 2002 and 2017 was 50 $52 billion. And that was for all of Amazon, not just AWS between 2002 and 2017. If we're talking about the dot com bubble. The dot com bubble's economics were actually fundamentally different. Fiber was much cheaper and also had more uses, more obvious uses. Let me put it really simply, I

[AI Industry Analyst]

know what you're getting at. But to put it really simply, when the.

[Ed Elson]

My point isn't wrong that it took nine years to get profitable for Amazon and there were many other dot com companies, it took many, many years to get profitable.

[AI Industry Analyst]

But Amazon was also not just a cloud compute company.

[Ed Zitron]

It was a. Basically went from a bookstore to a

[AI Industry Analyst]

store to a cloud compute company.

[Ed Zitron]

Added totally different business. Right?

[Ed Elson]

Yes.

[AI Industry Analyst]

And the point I'm making is the way this is being invested in doesn't

[Ed Zitron]

have the recovery story from the dot

[AI Industry Analyst]

com bubble because with.com bubble, the dark fiber that was laid and the interchanges

[Ed Zitron]

and the various bits of Lucent Telecom

[AI Industry Analyst]

stuff that was left abandoned, it didn't

[Ed Zitron]

cost a ton of money to operationalize it and its running costs were not incredibly expensive.

[Ed Elson]

Yes.

[AI Industry Analyst]

If data centers end up not being built, it's going to cost just as much money to finish them in 10 years as it will today.

[Ed Zitron]

And then on top of it, electricity

[AI Industry Analyst]

costs are likely to be more in

[Ed Zitron]

the future, not less and there will be less customers.

[AI Industry Analyst]

I understand where people get this from.

[Ed Elson]

Yep.

[AI Industry Analyst]

It's just a difficult, Difficult to square with me.

[Ed Elson]

Yeah, I think so. Those points I agree with and I think it's something that needs to be talked about more and I'm glad that you're bringing this up. My point, just to be clear about what my point was, the idea of these businesses, these business models not working yet to me doesn't seem to be a huge problem. But when you look at the amount that they are spending and when you look at just the sheer size of the spending to, to a lot of your points, I do wonder if when we extrapolate this, you know, 10, 15, 20 years down the line, whether it will be, whether it will look like the dot com boom and whether these Internet companies will be comparable. And I am increasingly thinking they won't. And something that I have been thinking, and this is something that Sam Altman has said, is I wonder if it looks like utility services. I mean, Sam Altman has said OpenAI is going to be a utility company. And we've seen this with utility companies before, where, you know, having a regular market doesn't really work because the economics don't make sense. And so you basically just have to have one company that's operating for a, a, a specific area. And so I guess what I'm getting at is the business model. I. I don't write it off entirely but it does seem that it will have to be different fundamentally from other businesses. And I wonder if you agree with that or if you think that they just won't work at all and they will therefore implode.

[AI Industry Analyst]

So if we think about the dot

[Ed Zitron]

com comparison and we break down the kind of businesses we had with the dot com bubble you had really.

[AI Industry Analyst]

It was really two bubbles.

[Ed Zitron]

It was the telecommunications bubble and the software bubble. The software bubble and the E commerce bubble. The pets.com the cosmos and such.

[AI Industry Analyst]

Their economics didn't make sense because they

[Ed Zitron]

didn't make them make sense. They just were Petstock Comm I think was spending like 250 bucks a customer or something insane like that.

[AI Industry Analyst]

Yeah but it was being invested and it involved the movement of physical goods and there were. Those are margins you can bring down.

[Ed Zitron]

As Amazon did build their own logistics network and also chewy.com is basically pets.com Cosmo now exists as Instacart like there are Webvan. Pardon me. I think it's Instacart now not the literal same company, the same business model.

[AI Industry Analyst]

In the AI bubble you really have

[Ed Zitron]

three different kinds of companies.

[AI Industry Analyst]

You have AI labs, you have AI

[Ed Zitron]

startups so wrappers and then you have

[AI Industry Analyst]

infrastructure infrastructure within AI data.

[Ed Zitron]

Sense of GPUs is one of the most commoditized businesses of all time.

[AI Industry Analyst]

It's a horrible business but it's also. There's really no difference between a core

[Ed Zitron]

weave and an ebius and an iron or a lambda.

[AI Industry Analyst]

They're all backed by Nvidia.

[Ed Zitron]

They're all feeding money from Jensen Huang to Jensen Huang.

[AI Industry Analyst]

Point is is the those companies do

[Ed Zitron]

not have really like there is no changing this business model. Even if I don't believe this.

[AI Industry Analyst]

There was a chip in the future that theoretically profitable Blackwell isn't Vera Rubin won't be Hopper wasn't. None of the Nvidia GPUs are clearly profitable for operators.

[Ed Zitron]

Otherwise we would have someone making a profit within the.

[AI Industry Analyst]

Within the telecoms and dot com bubble.

[Ed Elson]

Just to clarify what you mean there profitable to use those GPU chips and then generate a return from AI model.

[Ed Zitron]

Yes, exactly. More money than they have spent.

[Ed Elson]

Yes. Not but not Nvidia. Very profitable for Nvidia. Just to be clear.

[AI Industry Analyst]

Nvidia incredibly that's the thing Nvidia memory stocks. They're all profitable because they're pre selling a bunch of capacity that storage and

[Ed Zitron]

servers and all that.

[AI Industry Analyst]

But a vast majority like the dot bubble there was was also Much smaller than this.

[Ed Zitron]

As far as actual money being invested, venture capital was a. I think one round from Anthropic this year is bigger than all the venture capital that went in during the do. It's insane. And that's. I mean even with inflation it's still pretty close.

[AI Industry Analyst]

AI rapper startups are all dying and also the way that they are raising and dumping money is not something that's creating any useful residue.

[Ed Zitron]

Harvey for example, I think they've raised

[AI Industry Analyst]

$900 million and they make 200, 300 million ARR. Oh, those are stinky margins. But the point is it's not like Harvey is buying a bunch of stuff

[Ed Zitron]

and doing proprietary data or buying servers. Like happened a lot during the dot com bubble because AWS didn't exist so people were buying their own servers.

[AI Industry Analyst]

There's no useful residue. All of that money is being dumped directly into OpenAI and Anthropic.

[Ed Zitron]

Anthropic and OpenAI also do not have assets.

[AI Industry Analyst]

They don't have capex.

[Ed Zitron]

They have their models, they have their model weights which by the way Microsoft has full and complete access to OpenAI's model weights I confirmed today.

[AI Industry Analyst]

Yeah, and so they don't have their. They have talent, I guess they have research, but they don't really have.

[Ed Zitron]

They just have their models.

[AI Industry Analyst]

That's the only thing those companies die. There's not really any useful residue to pick up.

[Ed Zitron]

Perhaps someone could buy the assets of Anthropic even though Amazon and Google will

[AI Industry Analyst]

take them, but there's not really a thing to recover. And then you get to the core weaves Nebius irons of the world.

[Ed Zitron]

And again you've got a bunch of

[AI Industry Analyst]

old GPUs probably in data centers that haven't been built yet, which means that there's nothing really to pick up other than you pick up a cheap office chair or a cheap server.

[Ed Zitron]

Back in the dot com bubble you could do something with that.

[AI Industry Analyst]

What you're going to do with a gpu, especially when it's insanely expensive to run. I have a source at Oracle that told me it's about $6.30 an hour of COGS to run a B300, pardon me, GPU from Nvidia. These margins do not make sense. These costs don't make sense. I don't see how you pick up the pieces here. I don't see it at all. And indeed I don't think there's much

[Ed Zitron]

useful infrastructure left over.

[AI Industry Analyst]

And this is the big argument that people are making because people are saying

[Ed Zitron]

we're in a Bubble now.

[AI Industry Analyst]

We're in a bubble now. But don't worry, after the dot com

[Ed Zitron]

bubble people picked up stuff for cheap. What stuff?

[Monarch Advertiser]

Right.

[AI Industry Analyst]

We already have piles of unused GPUs. Supermicro had a multibillion dollar order canceled from Oracle. They have at least a billion dollars of 200 chips that no one wants to buy. We've already got the fallow inventory. Nothing's happening. I'm sure someone will do something, but nothing's going to be cheaper in 10 years.

[Ed Zitron]

Things work like you could pick up the servers and you could actually put them in.

[AI Industry Analyst]

Like there's ways that you could make

[Ed Zitron]

that useful from the dot com bubble.

[AI Industry Analyst]

It's not really this story here.

[Ed Elson]

I think the, the part where I started to disagree is, I mean, I'd be interested to hear what your views are on AI as a use case. Like we've seen what I believe to be sort of BS industries before. One of them in my view was the crypto industry that I was very critical of for a long time. That to me was people trying to sell us a solution to a problem that doesn't exist and then hammering it over and over and over again. And my view was this, this isn't really a problem, this isn't really helping people. There isn't much of a use case here with AI. I don't think of that the same way. I mean I see a lot of what you're saying, which is companies are almost trying to force AI down their employees throats because they just, they know or they believe that this is the most important thing that they have to do. Figure out how to use AI. And I think a lot of that is BS in a lot of ways. But at the same time I do find AI useful in certain areas of my life. I know that other people do as well. I know that OpenAI has a billion weekly actives. I know that nearly 9 in 10 developers are using AI to code. What percentage of them are being forced to do that? I don't know, but I doubt that the answer is all of them. My belief is that there is some value here. And the question is how much of it is real, how much of it is fake, and how do we figure out the proportions between the two? But sometimes I get the sense that you believe that there is no value, that it, that it won't work.

[AI Industry Analyst]

I think that there is. I've, I have very specifically said there

[Ed Zitron]

is value in coding. I've said that for years. That's classic misinformation there. Not from you it's, it's, it's a spread, it's spread from, it's spread all around by certain.

[Ed Elson]

No, I'm just, I'm hearing what you're saying and I'm, I'm trying to.

[Ed Zitron]

No, no, no, no, no. But you're not, you're not saying this, just being clear. You've been very, you've been very fair to me.

[AI Industry Analyst]

Point is, there is use, but all

[Ed Zitron]

of that use and that utility is subsidized.

[Ed Elson]

Yes.

[Ed Zitron]

So GitHub co pilot. I talk a lot about anthropic. We can talk about GitHub co pilot.

[AI Industry Analyst]

They were selling $15 for a dollar. Like that's fundamentally.

[Ed Zitron]

There was a person who posted. They've seen someone who spent 39 bucks a month. They spent 1500 dollars worth of tokens. Someone else 5 grand, someone else 3

[AI Industry Analyst]

grand, someone else 200.

[Ed Zitron]

This is direct money to Anthropic and OpenAI. This is not something where it's. Oh, they can claim they're profitable on inference. This is just straight up token boom.

[AI Industry Analyst]

Yeah, I would love to see how popular this was if people actually paid what it cost. Because I think that right now this all you can eat nonsense is inflating the value of it. Because if you were paying anthropic estimates

[Ed Zitron]

$13 a day is what someone will spend using Claude code.

[AI Industry Analyst]

That's a lot of money. Hundreds of dollars a month. I don't think most people get that

[Ed Zitron]

much value out of it.

[AI Industry Analyst]

If you had to, every little fun little turn you had with OpenAI, a couple cents each time you're spending several

[Ed Zitron]

dollars a day, I don't think people

[AI Industry Analyst]

would think it was quite as useful. Yeah, I also, the hallucination problem, the

[Ed Zitron]

incorrect information, it applies everywhere.

[AI Industry Analyst]

It's a very bad thing. But on a fundamental utility level, if this was being charged at cost, I actually think that every journalist should be

[Ed Zitron]

barred from using it in any way other than token based billing.

[AI Industry Analyst]

Because that's the only way you see the realistic cost here. Because, yeah, if you get this thing massively subsidized and you don't know the real cost, you're creating a completely fantastical environment.

[Ed Zitron]

100% to explain these things.

[Ed Elson]

Hundred percent.

[AI Industry Analyst]

And so. And people will argue, well, the cost of intelligence has come down. No, not at all. That is a very weird aberration where they've said, well, to do a task that you did two years ago, it's now 10x cheaper. If your dog's awake but your cat's asleep and it's like a Weird, weird series of asterisks like Barry Bonds baseballs. The thing is, we are yet to

[Ed Zitron]

actually meet the point when people really have to judge how valuable this is.

[AI Industry Analyst]

We are seeing it already. ServiceNow just blew through their AI yearly

[Ed Zitron]

token budget in what, five months. Uber burned through it in four.

[AI Industry Analyst]

We're seeing now the real costs start to come in. That's where my big argument is. And this stuff is not getting cheaper.

[Ed Zitron]

It's.

[AI Industry Analyst]

It isn't. They might be charging less for some models, but that is not necessarily helpful

[Ed Zitron]

if the models are burning more tokens. Killer Code had a great piece about this.

[AI Industry Analyst]

So in the end it comes down to, yeah, if this costs nothing, that

[Ed Zitron]

would be a completely different conversation.

[AI Industry Analyst]

But this costs everything. And the argument that people keep making

[Ed Zitron]

is that training costs will go away. No, they won't. When?

[AI Industry Analyst]

How? In fact, very simple question.

[Ed Zitron]

How does Anthropic become profitable?

[AI Industry Analyst]

And my God, if someone says they're profitable on inference. That comment came from three separate times when Anthropic's Dario Amade, Mr. Wario himself, has said, well, here's a stylized fact. Say you're 50% profitable on inference. It's always this weird duck and weave thing that is decidedly non GAAP and also entirely made up. Like he even said it was made up. So people have come up with this vision. The inference is profitable. If it was, 85% of Anthropic's revenue comes from API calls. Why is Anthropic losing billions of dollars a year then? And if the answer is training, okay, when does training go away? Why is training getting more expensive every year? The answer is training is a cost of goods. Yeah, it's. There is no profitability for any of this and you have to keep training the models because of model drift.

[Ed Elson]

I think these are all reasonable points. I would add that this as a concept isn't new. And this is sort of what. This is what venture capital is all about. And this is sort of taking it to another level. But if, you know, the idea of subsidizing a product that making it a lot cheaper and in some cases free, until it's reached a level of scale and a level of penetration that people are willing to pay for it, and then suddenly you can sort of turn the business on and it works. I think a good example of that would be Uber, for example, where, you know, we were all taking Ubers because it was subsidized and the VCs were essentially paying for our rides. We keep using it and using it and using it the network effects grow and then eventually everyone's using Uber and then eventually it turns into a real business that is profitable. Now to your point, but this would

[Ed Zitron]

be, this would be like if Uber paid for the cars, paid for the gas.

[AI Industry Analyst]

The gas was giraffe blood.

[Ed Zitron]

They pay for the customers clothes.

[AI Industry Analyst]

Like I'm, I'm being facetious, but like it is, the economics of Uber were also.

[Ed Zitron]

They didn't do.

[Ed Elson]

To a different degree.

[AI Industry Analyst]

Yes, well, also they didn't do the

[Ed Zitron]

same kind of subsidy.

[AI Industry Analyst]

They made a trip cheaper. Way cheaper.

[Ed Zitron]

Way, I remember those days, way cheaper.

[AI Industry Analyst]

But they didn't charge like they still charge, trip to trip, point to point, even this is. We are charging. You're paying $20 to drive 100 miles a month. Except in reality, each mile is costing

[Ed Zitron]

Uber 25 bucks or more. And there is no compelling argument as to how that changes.

[Ed Elson]

Yeah, I mean, it seems as though it's a bet on eventually when we flip that switch, the prize is going to be so gigantic that it ekes out what has been a gigantic cost. And to your point, the costs are gigantic and they're not really going down. I guess the question becomes, will that happen? I mean, there's a question of the certainty with which the number is going to be negative versus positive when we turn that switch on. Yeah, lots more that we can get into. I do have to wrap us up here, but I'm sure we will be discussing this another time. Ed Zitron is the author of the where's your Ed at? Newsletter. He is the host of the Better Offline podcast. Ed, I'm so glad we finally had you on the podcast. Lots more to get into. Thank you for joining me.

[Ed Zitron]

Thanks so much for having me.

[Ed Elson]

After the break, Trump meets with Xi Jinping. And by the way, we're heading out on tour at the end of the month. So for more info and to get tickets to a show near you, head to profgmarketstore.com.

[Monarch Advertiser]

Support for the show comes from Monarch. With the weather warming up, you're probably looking at more time out and about. But you can't really enjoy your time in the sun if you're completely caught up in how much money you're spending and how much you have left in your bank account. Thankfully, there's Monarch. Monarch is the personal finance app that tracks everything, accounts, investments, savings goals and spending. Get your first year of Monarch for half off, just $50 with promo code markets. Most apps only tell you what you've already spent, but with Monarch, you can set goals, map out big purchases and see if you're actually on track before it's too late to adjust. You can ask Monarch's AI assistant anything about your finances, like how much did I spend on travel last summer? Or can I afford this vacation without touching my savings? Plus, you can get a heads up on what's happening with your money with the AI weekly recap. It flags spending spikes, net worth shifts and upcoming expenses. Just use code marketsonark.com to get your first year for help half off at just $50. That's 50% off your first year at monarch.com with code markets. Support for the show comes from Anthropic. Some questions don't have clear cut answers,

[Claude/Granola Advertiser]

and when you're trying to work through

[Monarch Advertiser]

a difficult problem, it can help to have a partner to exchange ideas with and explore what's really going on underneath. That's where Claude can help. Cloud is the AI for minds that don't stop at good enough. It's a collaborator that actually understands your entire workflow and thinks with you. Just point it to a folder on your computer or add connectors like Google Drive and Gmail, describe what you need and it handles the rest, like organizing files, building spreadsheets, or drafting reports from scattered notes. You can queue up tasks and it comes back with finished work. So this isn't in the script. I linked my Gmail and I have just I found it so incredibly helpful. The other day I was trying to figure out I made an investment about seven years ago. I couldn't remember what I invested or what it was worth or how many shares I had in this private company. I asked Cloud to go to my Gmail and it found the thread from 2019 and almost instantly figured out exactly how many shares I had. I have found that enabling it to search my Gmail has really been a big unlock for problems worth solving. Get started with Cloud today at Cloud AI Markets. That's Claude AI Markets and check out Cloud Pro, which includes access to all of the features mentioned in today's episode. Cloud AI Markets. Support for today's show comes from Granola. You've been in enough back to back

[Claude/Granola Advertiser]

meetings to know how draining that can be. You're nodding along, contributing, trying to stay present, but you're doing everything you can not to fall asleep. Meetings can be a mess. Granola fixes that. Granola is an AI powered notepad built for the way real people actually meet. You simply take rough notes like you normally would in the background. Granola securely transcribes the meeting. Then it turns everything into Clean, structured, actually useful notes when the meeting ends. And here's the best part. Granola works through your device's audio, which means it integrates seamlessly into the video conferencing tools you already use. No setup, no awkward bots. It's just your normal meeting with superpowers. You need to do a better job. So if meetings are eating up your day, granola is a no brainer. You can try it totally free for three months. Just head to Granola AI Markets. That's Granola AI Markets. To get your time back, get three months free at Granola AI Markets.

[Ed Elson]

We're back with Profg Markets. President Trump just touched down in Beijing for his first visit in almost 10 years. He arrived with 17 CEOs, including Tim Cook, Elon Musk, Jensen Huang, and Larry Fink. Trump and Xi Jinping are sitting down to discuss trade AI rare earth, Taiwan and Iran. This meeting comes as the US And Iran remain deadlocked and the Strait of Hormuz is still closed, disrupting the globe. The US And China are also in the midst of a fragile trade truce that is set to expire this fall. So here to help us break down what we might see in this meeting, we're speaking with James King, host of the Prof. G Media's China Decode podcast and also senior research Fellow at Chatham House. James, good to see you. Thank you for joining us. What are we going to see in this meeting? What are you looking out for?

[James King]

Thanks a lot, Ed. Great to be with you. I think that obviously the two sides are hoping for different things. For Trump, I think this summit is all about three Bs and one I. The three Bs are beans, beef, and Boeing jets. Those are the big deals I believe that the Trump delegation is hoping for. And the I, of course, is Iran. I think that the White House is understood to be kind of needing Chinese acquiescence, if not assistance in pressuring Iran to unblock the Strait of Hormuz and try to end the war. To me, that's the kind of crystallization of what the US Side is looking for. The Chinese side. Well, these summits for China are all always about Taiwan. Beijing will be hoping that either Trump will say something big on Taiwan that moves the US Closer to China's position on Taiwan, and particularly whether Taiwan is an independent country, which Beijing strongly objects to, or or not. So Taiwan, I think will be there. And also I think that China will be hoping that access into the US Market will be smoother for lots of Chinese companies. And I think the last Thing is more of a kind of mood music thing, but an important one to me. The Chinese want to give Trump face so that the US in the future doesn't turn vengeful again against China. So to me, me, those are the, you know, the outlines of what either side is looking for. But, you know, these summits, they can surprise. There is an extraordinary meeting going to take place between Trump and Xi in the Temple of Heaven in Beijing. It's an amazing old shrine to the old kind of agricultural gods of China. So that's kind of amazing. I don't know, there could be surprises, but at the moment, I think those are the. Those are the contours.

[Ed Elson]

Why did he bring all these CEOs along with him? What are they going to do? What's Jensen Huang doing? What's Elon doing? Tim Cook, what is their role in all this?

[James King]

I think, yeah, as you rightly say, Stephen Schwarzman from Blackstone and Blackrock's Larry Fink, Elon Musk From Tesla and SpaceX, Tim Cook from Apple, they're all there. Jensen Huang from Nvidia is there as well. And Trump already said on the plane as he was coming over, quote, I will be asking President Xi, a leader of extraordinary distinction, to open up China so that these brilliant people can work their magic and help bring the People's Republic to an even higher level. My sense is that he's not that interested in bringing the People's Republic to an even higher level. I think he's interested in helping these companies win deals, win market access, and maybe stave off some of the actions that we've seen China begin to take against not just American companies, but all companies that would like to put sanctions on China or to decouple from China. So there's a very major commercial aspect to this whole trip as well, which

[Ed Elson]

seems striking to me, because it seems that at the start of this administration, it was all about kind of separating and kind of putting up the tariffs, figuring out a way to get ahead of China, sort of establishing we are two different opposing nations. I mean, the idea of bringing these CEOs, I mean, we even saw the chip restrictions which directly affected Jensen Huang and Nvidia. So the idea that he's bringing these guys along and I guess trying to reestablish a sense of a relationship and a trade relationship, I don't think that's a bad thing. I'm just a little confused as to why he's doing it. It seems like a 180 to me. Is that right?

[James King]

Yeah, that's Such a great insight, Ed. If I had to sum this up in one sentence, I would say that Trump is going to Beijing to rectify some of the problems that he himself created. As you rightly say, the U.S. the U.S. put 145% tariffs on Chinese exports at one point last year. Chinese exports of the US and then after China said that it was going to restrict exports of critical minerals. These are critical minerals that are needed to make US Weapons, that are needed by all the big US Tech companies to make the products that they sell. After China threatened, well, did that, and then, you know, engaged with the US In a series of talks and finally backed down, Trump then blinked and now the tariffs are back down to 47.5% on average. So to me, the key inflection point in this relationship, in this Trump administration's relationship with China has been China's invocation of these critical mineral export sanctions on the US Last October. And then Trump blinked the tariffs, the US Tariffs on Chinese exports came down. And this trip has sort of been in the offing ever since. This is a patch up work. This is an attempt to make nice to the Chinese. And this is why I was saying on the previous episode of China Decode. If you ask me, this is the first summit in US China history where the Chinese president has the upper hand and Trump is going there asking, you know, rather than the other way around.

[Ed Elson]

Yeah, it seems so striking and especially given a conversation that I had with your co host Alice Hahn last week where she pointed out that, I mean, when you think about the after effects of the Iran war, the real winner here in a lot of ways has been China. China has the upper hand in a lot of ways. And now we have Trump going to China in a position, in a probably a weakened position for both those critical minerals point that you point out there. But also when it comes to Iran and you've got inflation which is rising and you've got consumer sentiment plummeting and people saying they don't support this war, it seems like he's almost begging for something here, in which case he's gonna have to give them something in return, which makes me think maybe he will give them something when it comes to Taiwan.

[James King]

Well, yeah, I must say our thinking is very much aligned on this. In addition to what we've just been talking about, the commercial side of it, obviously Trump is looking for deals. He also needs China. On the issue of Iran. And I think that Iran is becoming a more and more pressing issue for the US There are all kinds of secondary effects that we're beginning to see. I really think that Trump needs to get done with this war with Iran and the route to that could lie significantly through China. Obviously. China is the biggest trade partner of Iran. It buys by far the biggest chunk of the Iranian oil exports. It has long longstanding relationships with Iran. It's got a 25 year economic agreement with Tehran, et cetera, et cetera. So I think that is also what Trump is aiming at. So, yes, I believe, I wouldn't quite say that Trump is going to Beijing as a supplicant, but I definitely think that Trump has a bigger ask from the Chinese than the Chinese have of Trump. And, and this is why I really do think this is historic, the first time that we've seen an American president go to China for a summit under those conditions. I mean, I've been covering these US China summits since the middle of the 1980s. This is remarkable. This really is remarkable.

[Ed Elson]

All right, James King, host of the China Decode podcast and senior research fellow at Chatham House. To hear more from the China Decode team, Alison James will be going live this Friday, May 15th at 10am Eastern on substack to break down the meeting between Trump and Xi Jinping. They will be joined by Kevin Hsu, founder of Interconnected Capital and writer of the Interconnected newsletter. To catch it live, become a subscriber of Prof. G [e-mail] you also get the China Decode newsletter every week and first notice on future live streams. James, thanks for joining us.

[James King]

Thanks so much, Ed.

[Ed Elson]

The producer Price Index was published yesterday. This is a measure of wholesale inflation, that is the prices that businesses are paying for domestic products. And the results were quite ugly. The index rose 6% from a year ago. That was higher than expected. And it was also the largest increase we've seen since 2022. Yes, 2022, the year we experienced the highest inflation in 40 years. And also the year the stock market fell more than 18% as a result. Now, going back to 2022, let's just remind ourselves why that all happened. For one, we had Covid, which blocked up global supply chains, which made it more difficult and more expensive to transport goods around the world. And two, Russia launched a war on Ukraine, which interrupted global energy flows and resulted in higher gas prices around the world. Emphasis on around the world, because inflation didn't just rise in America, it rose everywhere. Global inflation hit roughly 8% that year, which is why it doesn't really make sense when people say Biden is the Reason prices went up. No, prices went up because of two extremely large exogenous events that were actually out of our control. Now let's compare it to today. Why are prices rising in 2026? There are two reasons. Number one, tariffs. Last year we decided to issue indiscriminate tariffs on our global trading partners, which immediately raised prices for our own consumers and brought inflation up from 2.3% before Liberation Day to 3% in less than six months. And the second reason that prices are rising is of course, Iran. This year we decided to drop bombs on the nation that controls roughly a quarter of the global flow of oil. And as expected, fuel prices shot up, freight prices skyrocketed, and now gas prices in America are up roughly 30%. You will notice the big difference between inflation this year versus inflation four years ago. And that is this year's inflation is notably self inflicted. We chose it. If we had just sat around and done nothing, no tariffs, no wars, inflation would be roughly half where it is today. We would have hit the Fed's target of 2% and we would have been on track for rate cuts. But instead, the PPI is rising, which means the CPI will rise even further. And we are now looking at the possibility of of rate hikes. You could not have dreamed up a more self destructive economic policy than the one we are witnessing today. I might be our saving grace. We'll see. But when it comes to the economic decisions we've made, we have totally and utterly let ourselves down. First word? Own. Second word Goal. Okay, that's it for today. This episode was produced by Claire Miller and Alison, edited by Joel Patterson and engineered by Benjamin Spencer. Our video editor is Brad Williams. Our research team is Dan Shalon, Isabella Kinsel, Kristen o' Donoghue and Mia Silverio. And our social producer is Jake McPherson. Thank you for listening to Profg Markets from Profg Media. If you liked what you heard, give us a follow. I'm Ed Elson. Tune in tomorrow for our conversation with Professor Aswath de Modorin. Support for this show comes from Vanta. If you run a business, you don't need us to tell you that risk and regulation are rising. More and more customers expect clear proof of security before they'll commit. Building that trust is essential to closing deals, but it can also be costly, confusing and time consuming. Vanta says they can help. Vanta automates your compliance process to bring compliance, risk and customer Trust together on one AI powered platform. So whether you're prepping for a SoC2 or running an enterprise GRC program, Vanta keeps you secure and keeps your deals moving. And with continuous monitoring, Vanta helps with real time reporting and security reviews. You can share instantly. Instead of searching through audits and spreadsheets, you get a system working quietly in the background, keeping you compliant, reducing risk and helping your business grow faster. With confidence, you can get started at vanta.com markets. That's V A N T A dot com markets vanta.com markets

[Bloomberg Advertiser]

Some follow the noise. Bloomberg follows the money. Whether it's the funds fueling AI or crypto's trillion dollar swings, there's a money side to every story. Get the money side of the story. Subscribe [e-mail].