When the Trump administration cracks down on anthropology, who benefits?

Anthropy recently It has taken its two latest AI models offline Due to the Trump administration’s export control order, which has sparked widespread discussions about artificial intelligence policy and Digital sovereignty.

In the last episode of TechCrunch’s Stocks PodcastSean O’Kane, Rebecca Bellan, and I discuss why the administration has actually taken steps against Anthropics, and what this might mean for the broader AI ecosystem.

As Sean said, “Anthropic hasn’t had the best relationship with the Trump administration in a way that other leading AI labs have,” so Anthropic’s other competitors probably don’t need to worry about a similar crackdown.

But Rebecca also noted that their top cybersecurity experts “She signed an open letter asking Trump to rescind the orderThey say it is already dangerous to have to withdraw such advanced cybersecurity capabilities from network defenders in the United States.

We wondered if all this could be good publicity for Anthropic, especially since – as Rebecca put it – “everyone loves a bad boy.”

Keep reading for a preview of our conversation, which has been edited for length and clarity.

Rebecca Bellan: As I’m sure many of our listeners know, the US government basically forced Anthropic to pull its two newest models offline — Fable 5, and then there was also Mythos 5, which was available to existing Mythos users, (while) Fable 5 was more publicly available.

They sent a letter (last) on Friday that cited “national security concerns.” No one knows what those fears are. This report was not made public, they did not provide any details and told Anthropic that they had to ensure that these forms could not be used by any foreign nationals. So Anthropic was like, “Well, I think we should pull the models completely, because we don’t know when someone is a foreign national. A lot of our employees are foreign nationals.”

But in reality, (reports said) this was reported to the White House because of some researchers at Amazon who allegedly found a way to bypass Fable 5’s guardrails. Amazon CEO Andy Jassy has raised these concerns with the White HouseAnd it escalated from there.

Sean O’Kane: This all moved very quickly, especially on a Friday afternoon on a weekend. At the same time, the administration was ostensibly trying to negotiate some sort of treaty for the war it had begun in Iran.

Rebecca: Friday evening to us in New York. They love distraction.

Shawn: Let’s step back a lot for a moment. Anthropic hasn’t had the best relationship with the Trump administration in a way that’s different from other leading AI labs — I think there’s an element, at least, of that at play here.

Do you think this will have implications for those other companies? Do you think the Trump administration would be less inclined to turn off the tap on one of these competitors?

Anthony Ha: Part of the context here is that both reports and Analysis from independent security experts Point out that the actual security risks from Anthropic are not unique. So a lot of this seems to very much stem from parts of the Trump administration and the (mis)aligned anthropists very well. Whatever dangers there are, these things will blow out of proportion just because they can’t seem to make a civil phone call to each other.

If you’re another company – on the one hand, maybe it’s good for you, because you can say, “Okay, we’re not making these people mad at us and we can do whatever we want.” But it’s also not a great organizational scene to just say, “Boy, I hope they don’t get mad at us.”

Rebecca: On the one hand, it definitely seems vindictive – after the government called Anthropic a risk to the supply chain, and there’s this big lawsuit going on between them, it really seems like the White House is just looking for any excuse to hit Anthropic. And I feel this way not only because that was my initial reaction, but because of what a lot of cybersecurity researchers have said. They say this should never have led to an export control order. They have all signed an open letter asking Trump to rescind the order, and say it is already dangerous to have to take away such advanced cybersecurity capabilities from network defenders in the United States. Anthropic itself said that some of the same jailbreaks could have been found in many other AI models.

Sarcastically, it’s like: Well, are you just pausing Anthropy so other people can catch up with what Anthropy has been up to?

But at the same time, I also saw reactions that said (for example): This was kind of anthropic. They’re saying, “This is too dangerous for anyone to use, but not us. We’re the good guys.” They speak from both sides of their mouth. A week before Fable came out, they were (saying): “Hey, we need to slow down the AI, guys. This is getting really dangerous.” But then boom, “This is our craziest model ever, and it’s a super powerful model, go.”

Anthony: In some ways, this seems like a microcosm of a lot of discussions about AI, where people like Sam Altman and Jensen Huang say: “Hey, let’s try to lower the temperature. Why is everyone mad at us?” Well, you’ve spent the last two years saying you built this god machine that’s going to take jobs away from everyone. It’s not exactly a shock that people don’t feel good about this.

And there’s something about the way the Anthropists talk about the Mythos in particular, where they say, “This is the most incredibly powerful model ever, and it’s very dangerous to release it to the public.” And so on some level, (you say,) “Okay, well, let’s say we take this seriously then. That means there’s going to be an incredible level of scrutiny around this.”

And I’m wondering – it looks like Anthropic isn’t happy about this. I want to be careful about not overestimating how beneficial this is for them. But we’ve also published some stories about Ramp’s analysis Highlighting the fact that the recent major spat between Anthropic and the Trump administration has been beneficial for the companyat least in some respects. Cloud downloads are up. I think a lot of people might have thought about ChatGPT as well the chat bot, the The AI ​​assistant from before, suddenly they were looking at Claude as perhaps the most responsible, most “resistant” person.

In the same way, (while) anthropists are very concerned about this, this may make their models seem more robust again.

Rebecca: definitely. “We are too dangerous.” Everyone loves a bad boy, right? Everyone says: “It’s the strongest model, even Trump says that. Of course, I have to get my hands on it.”