Anthropic’s reported AI safety bill would limit federal intervention

Politico reported Anthropic pitched a draft with high harm thresholds; its evaluator process and state-law approach differ from the public framework.