# AI firms may use safety rules to block cheaper Chinese models

Markdown copy of https://realalpha.app/s/ai-firms-may-use-safety-rules-to-block-cheaper-chinese/ on Real Alpha, the best minutes from long interviews with the people building AI. Cite and link the page, not this copy.

- Speaker: [Steve Eisman](https://realalpha.app/people/steve-eisman/) (Host) · The Real Eisman Playbook
- Show: The Real Eisman Playbook
- Episode: [Why Young Americans Feel Locked Out with the Lemonade Stand Podcast | The Real Eisman Playbook Ep 82](https://realalpha.app/v/pTFWpojm1dU/), Nov 2, 2026
- Moment: 23:41 to 26:33 of the episode (3 min), playing on https://realalpha.app/s/ai-firms-may-use-safety-rules-to-block-cheaper-chinese/; the episode from there: https://realalpha.app/v/pTFWpojm1dU/?t=1421

## Quote

> “And the Chinese open weight models are literally **one-tenth the price**.”
> — [Steve Eisman](https://realalpha.app/people/steve-eisman/) at [25:00](https://realalpha.app/v/pTFWpojm1dU/?t=1500) · The Real Eisman Playbook

## Takeaways

1. Eisman says OpenAI and Anthropic face cheaper Chinese competition.
2. He puts Chinese open-weight models at one-tenth the price.
3. He argues safety warnings could support rules excluding those models.

## About this segment

Steve Eisman argues OpenAI and Anthropic fear Chinese open-weight models that cost one-tenth as much, so they press for existential-risk regulation. He says the rules could keep cheaper rivals out of the U.S.

From “Why Young Americans Feel Locked Out with the Lemonade Stand Podcast | The Real Eisman Playbook Ep 82”, Nov 2, 2026. [Watch the full episode](https://realalpha.app/v/pTFWpojm1dU/)

## Transcript

Transcribed by machine (Whisper) from the episode’s audio, so a word may be off here and there.

[23:47](https://realalpha.app/v/pTFWpojm1dU/?t=1427) Do you know what the problem is? Yeah. Everybody freaking watch Terminator. and you know what I'm actually being perfectly serious it's funny but I'm being perfectly serious by the way I re-watched Terminator 1 Terminator 2 still good Terminator 2 is a classic Terminator 2 is a classic Unbelievable. But even Terminator 1, which was like a low-budget film, is amazing. But I literally think part of the problem is—well, let me backtrack for a second.

[24:16](https://realalpha.app/v/pTFWpojm1dU/?t=1456) Sure. Growing up, I probably read more science fiction than any human being should ever be allowed to read science fiction. I mean, I challenge anyone in terms of the length of the science fiction list. It's not possible. Is she a nerd? But the people of Silicon Valley, they've read almost as much. There's one big difference. They take it totally seriously. Oh, yeah. They really take it serious. They think it's like a roadmap.

[24:43](https://realalpha.app/v/pTFWpojm1dU/?t=1483) They really do. And so my thesis about this whole insanity is that OpenAI and Anthropic right now are petrified. And they're petrified because they have created a business that produces a product that's very expensive. They're LLMs. Yeah. And the Chinese open weight models are literally one-tenth the price. Right. And so they are literally petrified that they're going to lose massive market share and have to cut costs, have to cut price.

[25:16](https://realalpha.app/v/pTFWpojm1dU/?t=1516) Yeah. So what they've done is that what they're trying to do is manufacture a crisis. And the crisis is, if you don't regulate us, planet Earth will end. So you better regulate us. Then when they'll regulate them, they'll figure out a way to get the regulators to throw all the Chinese open-weight models out of the country. That's what this is all about. Nothing more. Nothing more than that. What makes you—I don't doubt you when you say that.

[25:43](https://realalpha.app/v/pTFWpojm1dU/?t=1543) I feel like that is a very realistic evaluation of the situation. Why are you so confident about it? Because LLMs don't think. But this whole thing that LLMs are very close to artificial general intelligence, I think it's just wrong. Now, maybe one day we'll get there and we'll have to worry about Terminator, Skynet. When we're all on Arrakis mining the spice, we'll look back on this conversation.

[26:13](https://realalpha.app/v/pTFWpojm1dU/?t=1573) Right, that's the moment, the turning point. Turning point, exactly. So I just think it's nonsense. I just think it's utter nonsense. I mean, we can't know. I guess none of us know, Rodney. No, we don't. But you know what? You know what the thing is? I said this on a podcast earlier.