Anthropic’s CEO stuns Davos with Nvidia criticism

Last week, the US government decided, after reversing an earlier ban officially approved selling Nvidia’s H200 chips, along with a chip line from AMD, to approved Chinese customers. They may not be the shiniest and most advanced chips from these chipmakers, but they are powerful processors used for AI, making their export controversial. And on Tuesday at the World Economic Forum in Davos, Dario Amodei, CEO of Anthropic, said discharged both the administration and the chip companies about the decision.

The criticism was especially striking because one of those chipmakers, Nvidia, is a major partner and investor in Anthropic.

“The CEOs of these companies are saying, ‘It’s the chip embargo that’s holding us back,’” Amodei said incredulously in response to a question about the new rules. The decision will come back to bite the US, he warned.

“We are many years ahead of China in our ability to make chips,” he told Bloomberg’s editor-in-chief, who interviewed him. “So I think it would be a big mistake to ship these chips.” Amodei then painted an alarming picture of what is at stake. He spoke of the “incredible national security implications” of AI models that “essentially represent cognition, which is essentially intelligence.” He compared future AI to a “country of geniuses in a data center,” saying he would have to imagine “100 million people smarter than any Nobel laureate,” all under the control of one country or another.

The image underlined why he thinks chip exports are so important. But then came the biggest blow. “I think this is insane,” Amodei said of the government’s latest move. “It’s a bit like selling nuclear weapons to North Korea and… [bragging that] Boeing made the housings.”

That sound you hear? The Nvidia team screamed into their phones.

Nvidia is not just a chip company. While Anthropic runs on Microsoft and Amazon and Google servers, only Nvidia provides the GPUs that power Anthropic’s AI models (every cloud provider needs Nvidia’s GPUs). In addition to being at the center of everything, Nvidia recently announced that it is investing in Anthropic for up to $10 billion.

Just two months ago, the companies announced that financial relationshipalong with a “deep technology partnership” with cheerful promises to optimize each other’s technology. Fast forward to Davos and Amodei compares his partner to an arms dealer.

Maybe it was just an unguarded moment; it’s possible he got swept up in his own rhetoric and blurted out the analogy. But given Anthropic’s strong position in the AI ​​market, it seems more likely that he felt comfortable speaking with confidence. The company has raised billions, is valued at hundreds of billions, and its coding assistant Claude has built a reputation as a highly beloved and high-quality AI coding tool, especially among developers working on complex, real-world projects.

It’s also entirely possible that Anthropic is genuinely afraid of Chinese AI labs and wants Washington to take action. If you want to get someone’s attention, nuclear proliferation comparisons are probably a pretty effective way to do that.

But what is perhaps most remarkable is that Amodei could sit on stage in Davos, drop such a bombshell, and walk off to another meeting without fear of negatively impacting his business. News cycles continue, sure. Anthropic is also currently on solid footing. But it does feel that the AI ​​race has become so existential in the minds of its leaders that the usual constraints – investor relations, strategic partnerships, diplomatic niceties – no longer apply. Amodei doesn’t worry about what he can and cannot say. More than anything else he said on that stage, that fearlessness is worth paying attention to.