Anthropic CEO Dario Amodei says he does not support open-weight AI ban

Anthropic CEO Dario Amodei said this week he's "never advocated" for banning open-weight AI models, rejecting accusations that his company wants to shield its closed-model business from competition. His statement comes after a shock Chinese open-weight model called Kimi K3—which approached U.S. frontier performance at a fraction of the cost—triggered alarm in Silicon Valley and spurred Nvidia, Microsoft, Meta, Google, and OpenAI to sign a letter urging Washington not to restrict open-weight technology. Anthropic notably has not signed on. Instead of a ban, Amodei is proposing three narrower policies, according to Axios: tighter controls on advanced chips flowing to authoritarian governments, a crackdown on training models on outputs from more advanced systems, and mandatory safety testing for capable models. He argues that outright bans wouldn't address his core national security concerns. Amodei acknowledged such restrictions would shelter U.S. labs from Chinese competition—but said that's never been his goal.

Listen to this story

Hear this and more stories in a personalized audio briefing.

Open The Chonkerton