Anthropic Pitches Draft AI Safety Bill to US Lawmakers With Strict Intervention Thresholds

Anthropic has quietly presented a draft AI safety bill to Capitol Hill lawmakers, Politico reported on Friday, citing three people with knowledge of the effort. The proposal focuses on frontier AI models and would establish specific thresholds at which the federal government could intervene to block a model's release or use.

Under the bill, government intervention would be permitted only if a model poses a foreseeable, imminent risk of killing or seriously injuring more than 50 people, or causing more than $1 billion in damages. The threat would have to stem from one of three scenarios: the creation or release of a biological weapon, a model evading its developer's controls, or a cyberattack. Independent evaluators registered with the Commerce Department would be tasked with periodically assessing labs and their frontier models. Anthropic and the Department of Commerce did not immediately respond to requests for comment.

The disclosure rattled markets, with Bloomberg noting that AI-related stocks dragged as the news contributed to broader jitters over AI regulation. The proposal marks a notable move by a leading AI developer to shape federal legislation on its own terms, arriving as Congress remains largely without comprehensive AI oversight rules despite repeated calls for action from industry figures and safety advocates.