Anthropic Proposes AI Safety Law to Block Dangerous Models

Artificial intelligence company Anthropic has submitted a draft bill to US lawmakers proposing stricter safety regulations for advanced AI systems. The proposal would grant federal authorities the power to prevent deployment of models deemed capable of causing catastrophic harm.

Under the proposed framework, federal regulators could block an AI model's release if it poses an imminent threat of killing or injuring more than 50 people or causing damage exceeding $1 billion. These powers would apply specifically to risks involving biological weapons, cyberattacks, or AI systems operating beyond human control.

Anthropicis draft legislation would require AI developers to conduct mandatory safety assessments before deploying advanced models and disclose their testing findings publicly. Independent evaluators would be permitted to conduct regular reviews of AI systems. An Anthropic spokesperson stated the proposal reflects recommendations from the company's Advanced AI Framework, which has been advocated to federal and state lawmakers since June. The company believes Congress should introduce safeguards addressing emerging technological threats while establishing clearer accountability for AI developers.