Anthropic to Revise Leaked Federal AI Safety Bill
Anthropic told US lawmakers that it will revise parts of its leaked draft federal AI safety bill after criticism from the AI safety community. Critics said the overall draft was weaker than the FRONTIER Act and identified weaknesses in evaluator standards and model-blocking rules. The draft would allow federal intervention in narrowly defined cases, preempt some state AI laws, and establish independent evaluators who register with the Commerce Department. Anthropic said its preemption wording was insufficiently precise, that risk thresholds should be reconsidered, and that it would provide further proposals on independent evaluators. The revised text and timing were not specified.
The proposal concerns federal authority to block AI models, preemption of some state AI laws, independent evaluation requirements and the thresholds for government intervention.

Anthropic has told US lawmakers it will rewrite part of its own draft AI safety bill. The leaked text drew days of criticism from the AI safety community. The confidential draft was published by Politico on Friday. Anthropic has been circulating it on Capitol Hill.
Critics say the bill is weaker than the main AI safety bill already before Congress. They say it would make a dangerous model harder to stop, not easier. That is an awkward position for a company that built its name on pushing for tough AI rules.