Anthropic CEO rejects openweight AI ban urges testing

The debate over regulating openweight artificial intelligence models intensified this week as Silicon Valley leaders outlined competing policy visions, with Anthropic CEO Dario Amodei explicitly rejecting calls for a blanket ban while advocating for targeted measures to address national security risks. Amodei's statement, published on Monday, July 27, 2026, sought to counter industry speculation that his company supports prohibiting openweight AI systems, which allow users to download and modify model code.

In his post, Amodei stated unequivocally that Anthropic has never advocated for a ban on openweight models as a category. He described such models as a public good when they do not possess dangerous capabilities, providing value to businesses, developers, and researchers. His position came days after a coalition of major technology companies, including Nvidia, Microsoft, Meta, and OpenAI, released a letter urging policymakers to avoid broad, premature restrictions on openweight AI. Anthropic notably did not sign that letter, a decision Amodei defended by saying he disagreed with its assertion that openweight models make it easier to develop safety safeguards.