Anthropic chief rebuts claims of supporting open-weight AI bans after industry letter

Anthropic’s chief executive Dario Amodei issued a public response after the company was not listed among signatories of a letter promoting open-weight AI models, which included Nvidia, Palantir and others. In his statement, Amodei emphasized that Anthropic has never supported a ban on open-weight models and questioned the claim that such models inherently make safety easier. He described non-dangerous open-weight models as a public asset that benefits developers, researchers, and businesses at modest computational cost.

Although he concurred with many points of the letter—such as broadening AI access and fostering competition—he warned that open models might also empower malicious actors. Amodei proposed focusing on restricting powerful chips from authoritarian regimes, preventing large-scale model distillation, and mandating safety evaluations for all capable models, whether open or closed. The response followed Palantir CEO Alex Karp’s interview on one outlet, where Karp argued that resisting competition would weaken the tech sector and highlighted the benefits of building open-weight AI platforms in the United States.