Anthropic proposes AI safety bill allowing model blocks if 50+ deaths are possible

Anthropic has put forward a proposed AI safety bill that could permit blocking a model where it might cause 50 or more deaths. However, the policy reportedly would bar federal agencies from halting a model's release without first obtaining a judge's approval. Critics of this framing may see the bill as offering less protection than its headline threshold suggests. The article presents the proposal as one that appears to guard against runaway AI while limiting regulators' direct power to intervene.