Anthropic Proposes AI Safety Bill Giving US Government Power to Block High-Risk Models

Anthropic has proposed draft federal legislation to US lawmakers that would create an AI safety framework giving government power to block deployment of AI models posing severe public safety risks, including threats causing over $1 billion in damage or serious injury to over 50 people. The proposal includes independent technical evaluations, mandatory risk reporting, and court-reviewed enforcement to manage high-risk AI systems.

This Wire brief sits within Fusion42's coverage of AI & ML, and 3 sources have reported it between 17 Oct 2026 and 19 Oct 2026.