Zvi Mowshowitz

Things Zvi Says on Podcasts

AI analyst and rationalist blogger who writes weekly roundups on language models, their uses and harms, and catastrophic AI risk, alongside posts on housing, education and sports. Member of the Magic: The Gathering Pro Tour Hall of Fame, founder of the policy think tank Balsa Research and co-founder of MetaMed. Publishes Don't Worry About the Vase on Substack and WordPress. Has appeared on The Cognitive Revolution, 80,000 Hours Podcast, Odd Lots and EconTalk.

Where to Find Them

Zvi Mowshowitz writes Don't Worry About the Vase (Zvi) . They have also been a guest on The Cognitive Revolution" | AI Builders, Researchers, and Live Player Analysis (17 times) , Complex Systems with Patrick McKenzie (patio11) (3 times) , 80,000 Hours Podcast (2 times) , Odd Lots (2 times) , Future of Life Institute Podcast , NonZero Newsletter , ChinaTalk , EconTalk , Risk of Ruin and What's Your Problem? .

Recently: “Reading the Anthropic S-1, with Zvi Mowshowitz” on Complex Systems with Patrick McKenzie (patio11) (November 2026); “AI #192: Anthropic Is Going Public” on Don't Worry About the Vase (Zvi) (October 2026); “Anthropic Files Its S-1” on Don't Worry About the Vase (Zvi) (October 2026); “Anthropic Versus Anthropic” on Don't Worry About the Vase (Zvi) (October 2026); “Gemini 4 Argon: The Model and Safety Cards” on Don't Worry About the Vase (Zvi) (October 2026); “AI:AM Highlights: Clean Neurons, Black October, the GPU Credit Crunch, and Buying the Supply Chain” on The Cognitive Revolution" | AI Builders, Researchers, and Live Player Analysis (October 2026).

What They Said

“Anthropic has been incredibly conservative, in my opinion, on its forward commitments and its acquisition of compute. OpenAI has been much, much more aggressive. And still, OpenAI proved to be more conservative than it would have been optimal to be. … The cost of not having the compute available when you go really well is very high. … The only way this works out badly for them is if the price of compute actively goes down substantially, so they can't resell it, and there's no demand for it. And that is a very strange world that I do not expect.” — Zvi Mowshowitz, Complex Systems with Patrick McKenzie (patio11)

Patrick had framed Anthropic's roughly $800 billion of forward compute commitments as equivalent to debt that could sink the company if demand slows. Zvi says the real mistake across the labs was committing too little, not too much.

Complex Systems with Patrick McKenzie (patio11) · 2026-11-06 Permalink → Listen →
Complex Systems with Patrick McKenzie (patio11) Around 33:33 into the episode
Zvi Mowshowitz

…any more than we should charge for Dario's stock going up in value and therefore having been a more valuable thing to give to Dario. There's a very silly way to do accounting. And then the actual expenses that we have to look forward to are Anthropic has compute commitments of roughly $800 billion.

Patrick McKenzie

…to having $800 billion of debt, just like it would be if you had $800 billion of real estate leases. And that could cause, quote unquote, the bankruptcy of the company if demand growth slows or yada yada yada. I have reasons for not thinking that, but what's your perspective on the forward commits?

Zvi Mowshowitz

…slash how dead they would be as a company in terms of their like plausible amount of quality future if Grove had in fact stalled for various reasons, that they were now falling out of the top tier and were not able to make these sales. That didn't happen. But if it had, it is not clear to me that not having bought enough compute would have saved things that were that valuable. Whereas like the cost of not having the compute available when you go really well is very high. Obviously, there are reasons to want to be risk reverse, but I do think they played it too conservatively, even though from an outside perspective, you know, oh, 800 billion dollars are forward commits, that's completely insane. That's over a number of years. Compute is growing rapidly. AI is growing rapidly. The only way this works out badly for them is if the price of compute actively goes…

Patrick McKenzie

Hideyo, folks. So a bit of a post-interview rift on this topic. There's an old saw: if you owe the bank $900, you have a problem. If you owe the bank almost $900 billion, the bank has a problem. What does a future look like in which the buyer is unhappy about the $900 billion in forward commits?…

Zvi Mowshowitz

Yeah, I think of lab compute spend as roughly divided in half between a combination of internal use, pre-training, and post-training on the one hand and serving the models on the other.

Patrick McKenzie

Yeah. And so hypothetically in the future, if demand for inference goes up, keeping compute constant, you can adjust for that by saying, okay, well, we'll accept. Some slowdown with regards to our next model release, or like shoot for less in terms of adult-down capabilities, or yada yada. And if…

Speaker names come from our own diarization. The position is estimated from where the line sits in the episode.
“Here they call recursive self-improvement 'closing the loop,' to be more business friendly. But they say their entire business model is they want to close the loop. They are worried that if someone else closes the loop before they close the loop, they will be out of the loop and they will lose. So they intend to close the loop as fast as possible and as much as possible.” — Zvi Mowshowitz, Complex Systems with Patrick McKenzie (patio11)

Zvi opens with his key takeaways from Anthropic's S-1, starting with how it treats existential risk, alignment and the pace of AI progress. He argues Anthropic is trying to appeal to investors and warn about danger in the same document. This is his reading of the filing's language on self-improving AI.

Complex Systems with Patrick McKenzie (patio11) · 2026-11-06 Permalink → Listen →
Complex Systems with Patrick McKenzie (patio11) Around 05:55 into the episode
Zvi Mowshowitz

I do buy individual stocks. I own Google and Amazon relevantly to Anthropic, but I deliberately did not invest directly in Anthropic because of potential conflict of interest concerns.

Patrick McKenzie

And nothing that you hear on this podcast is investment advice. Okay, so let's start it out. So you recently wrote something of your own perspective on the S1, which had some key takeaways. Do you want to walk people through those key takeaways?

Zvi Mowshowitz

Yeah. So as usual, when I look at anything from Anthropic, I want to see their attitude towards exit risk, alignment, whether or not they're going to potentially kill everyone. So in this case, they mention exit risk only in passing, only in the passive voice. They're clearly trying to have it both ways. When they talk to the investors, they want to say, oh, look at our big mission, but our big mission is to build the super profitable hypercomputing thing. And then other times they want to say our genius in the data center actually might kill everyone. And these two things have to coexist. Similarly, pacing the frontier is something they do talk about here. But they make care to investors. Don't worry. When we said pacing the frontier, we did not mean not make incredibly fast AI progress. We did not mean voluntarily let someone else go first. We just meant it would be great if everybody could collectively agree to go only at super speed and not at ludicrous speed. But they will do the embedded evaluators. Similarly, we're all worried about recursive self-improvement. And here they call that closing the loop to be more business friendly. But they say…

Patrick McKenzie

I feel like there is an interesting tension around S1s because we largely have this culture in the tech industry of necessarily like Zuck speaks for Facebook and Facebook speaks for Zuck. And we have complicated things with shared classes that enable us to be relatively durable over a period of…

Zvi Mowshowitz

Yeah. Obviously, it's a compromise between the bankers who would rather not mention this at all, because if everyone has killed you will not, in fact, be sued for securities fraud. There will be no one to sue you. And the, you know, Dario and others at the company presume we wanted to use not only…

Patrick McKenzie

And there's a liability sort of thing there where, you know, lawyers will always advise you from like, don't say anything that you don't need to say, anything that you don't need to say to get the transaction, deal, testimony, et cetera, done just creates risk needlessly. And so if you feel…

Speaker names come from our own diarization. The position is estimated from where the line sits in the episode.
“They still seem like pretty the most responsible lab, but that's rating on a curve. And I don't like rating on curve because reality doesn't have a curve. Reality just decides what happens.” — Zvi Mowshowitz, The Cognitive Revolution" | AI Builders, Researchers, and Live Player Analysis

Asked how to understand Anthropic now, the AI writer says it has a much stronger safety culture than its rivals but one that is not adequate to the moment. He argues its policy team acts like a normal corporation a remarkably large share of the time.

The Cognitive Revolution" | AI Builders, Researchers, and Live Player Analysis · 2026-10-24 Permalink → Listen →
The Cognitive Revolution" | AI Builders, Researchers, and Live Player Analysis Around 1:27:45 into the episode
Speaker 2

…needs to happen. All the alignment work still needs to happen. The mechanics stuff is great. We should absolutely do it. The biggest thing is, you know, we're going to have to meet the moment. And we're going to have to, at minimum, do something that counts as some form of facing the frontier.

Nathan Labenz

Nathan Calvin of ENCODE had publicly listed nine anthropic decisions he saw as putting the race ahead of safety. And Dario Amadei had replied at length. I asked Zvi Mowshowitz we should understand Anthropic now.

Speaker 2

…information, so many different bits of information that we should be worried about this. I've also, you know, recently from various employees that I talked to at various points gotten some rather alarming views about from my perspective on what it would take to solve alignment problems, what it would take to make the future go well. And all of that plays into being very highly uncomfortable. They still seem like pretty the most responsible lab, but that's rating on a curve. And I don't like rating on curve because reality doesn't have a curve. Reality just decides what happens.

Nathan Labenz

That's the week. Thanks for listening to the AI in the AM weekly highlights. The narration was my clone voice. Please tell us what worked and what didn't. See you in the morning. If you're finding value in the show, we'd appreciate it if you'd take a moment to share with friends, post online, write…

Speaker names come from our own diarization. The position is estimated from where the line sits in the episode.

Collections They Appear In