Anthropic's safety bill would let the US block an AI model if there's imminent risk of 50-plus deaths
The draft proposal would limit the circumstances in which the government could block the release of an advanced model and preempt some state AI laws.
Anthropic, an artificial intelligence company, is proposing a new AI safety bill that would allow the US government to block the release or use of an AI model under specific circumstances. The bill, obtained by POLITICO, outlines that the government can intervene if there is a foreseeable, imminent risk of more than $1 billion in damages or if the model poses a threat of killing or seriously injuring more than 50 people.
This legislation is based on Anthropic's Advanced AI Framework and is aimed at shaping the rules governing frontier AI technology. The bill emphasizes the importance of evaluating AI labs and their models to ensure they meet safety standards, with independent evaluators reporting any imminent risks to the Commerce Department within 72 hours.
However, no other US department, agency, or officer can suspend the model's development, deployment, or internal use. While the bill mirrors some existing California AI laws, it also aims to preempt similar state laws to create a unified national standard. The proposed legislation has faced mixed reactions, with some safety-focused groups and lawmakers advocating for a larger role of the government in reviewing AI models, while others push for a complete pause in advanced AI development and a ban on superintelligence.
Written by urgent.news from Business Insider's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
