Anthropic has put forward a draft proposal to U.S. lawmakers that could lead to the suspension of AI models deemed to pose significant risks. The draft legislation, which has not been adopted, defines those risks to include financial damage exceeding $1 billion or incidents causing 50 or more deaths or serious injuries. Concrete numeric thresholds of that kind are significant because they would give any future oversight regime measurable triggers for action rather than open-ended safety judgments.

Under the proposed oversight framework, the Commerce Department would be empowered to seek federal court orders to shut down AI models tied to specific threats, such as biological weapons or cybersecurity breaches — meaning any suspension would require judicial approval rather than unilateral action. Notably, the framework originates from an AI developer itself rather than solely from lawmakers, placing Anthropic directly inside Washington's broader effort to define how high-risk AI models should be overseen.

The proposal has registered in prediction markets that track the AI industry's competitive landscape. Prediction markets allow traders to buy and sell contracts tied to real-world outcomes, and their prices are widely read as crowd-s probability estimates. A market asking which company will hold the best AI model by the end of October 2026 now prices the likelihood of Anthropic holding the leading model at 21.5%, down from 24% roughly 24 hours earlier. The shift suggests market participants are weighing the regulatory environment and its implications for Anthropic's future development.

Taken together, the draft bill could signal increased scrutiny of, and potential intervention in, the AI sector by U.S. authorities — a dynamic that, according to the reporting, may introduce significant operational risks for AI developers, with Anthropic among the most closely watched.

Attention now turns to whether the U.S. Congress moves forward with adopting the proposed legislation. As a draft, the text carries no legal force; it would need to be formally introduced, clear both chambers of Congress, and be signed into law before any of its provisions could take effect. Regulatory action under such a framework could further influence market perceptions and the competitive landscape. Official statements or policy shifts from key lawmakers or the Commerce Department could provide additional clarity on how the proposal would affect AI developers such as Anthropic.