Here is what the page “Anthropic Proposes Federal AI Safety Bill: US Could Block Models With Imminent Risk of 50-” contains. Anthropic has drafted a federal AI safety bill that would empower the US government to block the release of advanced AI models if there is an imminent risk of 50 or more deaths. The proposal, reported by Politico, outlines a narrow trigger for intervention, aiming to address catastrophic risks while limiting regulatory overreach. It would also preempt some state AI laws, creating a unified national standard.
The bill focuses on frontier models with the potential for severe harm, such as enabling weapons of mass destruction or major cyberattacks. Anthropic's move comes amid a fragmented state-level regulatory landscape, with California's SB 1047 being a key point of debate. The company argues that a federal framework is needed to avoid a patchwork of laws and to ensure safety without stifling innovation. The full version stays on the original page, which this listing opens directly.