At the World Economic Forum in Davos, Anthropic CEO Dario Amodei delivered a striking and unexpected critique of both the U.S. government and major chipmakers, including Nvidia, in response to the administration’s decision to approve the sale of high-performance AI chips—such as Nvidia’s H200 and certain AMD processors—to approved Chinese customers. The move, which reversed an earlier ban, has sparked debate over national security and the future of AI competition. Amodei, whose company relies heavily on Nvidia’s GPUs to power its AI models, expressed disbelief at the administration’s stance. He challenged the narrative often cited by tech leaders—that chip embargoes are hindering U.S. progress—arguing that the U.S. remains years ahead of China in chip manufacturing capability. “We are many years ahead of China in terms of our ability to make chips,” he said during an interview with Bloomberg’s editor-in-chief. “So I think it would be a big mistake to ship these chips.” He warned of the profound national security risks posed by allowing advanced AI infrastructure to reach foreign hands. Amodei described future AI systems as “essentially cognition, that are essentially intelligence,” likening them to a “country of geniuses in a data center”—a digital army of intellects, potentially controlled by a single nation. The analogy underscored his concern that exporting powerful AI chips could accelerate China’s AI development in ways that threaten global stability. The most jarring moment came when he compared the decision to “selling nuclear weapons to North Korea and bragging that Boeing made the casings.” The remark stunned the audience and sent shockwaves through the tech industry, especially given that Nvidia is not just a supplier but a strategic investor in Anthropic, with a potential $10 billion commitment announced just two months prior. Despite this deep partnership, Amodei appeared unfazed, suggesting that the stakes of the AI race have become so existential that traditional business diplomacy no longer applies. His comments reflect a growing sentiment among leading AI figures that the race is not just about innovation, but about who controls the future of intelligence itself. Anthropic’s strong position—backed by billions in funding, a high valuation, and a widely respected AI assistant in Claude—gives Amodei the freedom to speak with such candor. He isn’t worried about immediate fallout, and the news cycle will inevitably shift. But the real significance lies in the message: AI leaders now operate with a sense of urgency and moral weight that transcends corporate alliances. Amodei’s fearless rhetoric at Davos signals a new era in tech leadership—one where national security, ethical responsibility, and strategic competition are no longer separate concerns, but deeply intertwined. In that context, the line between partnership and peril has become dangerously thin.