Anthropic lobbies for bill limiting AI model releases at high-risk threshold
What happened
Anthropic is lobbying for an AI safety bill, reported October 18, that it says reflects proposals in its Advanced AI Framework and that it has pushed to federal and state lawmakers since June. The proposed bill would let the government block a model's release only when it risks $1 billion in damage or more than 50 deaths or injuries. Independent third parties registered with the Commerce Department would assess labs and models periodically; labs can pay for assessments as long as that doesn't affect the assessors' independence. The report covers a proposed bill and lobbying activity.
Written by AI from the coverage · updated 1 hour ago
Coverage
Follow the reports to see the story from different sides.
- Hacker News front pageAnthropic's proposed AI safety bill only kicks in if 50 deaths or $1B in damages
Anthropic正游说推动一项AI安全法案，仅当模型存在造成$1 billion损失或超过50人死亡或受伤的风险时，政府才能阻止其发布。据Politico报道，在商务部注册的独立第三方将定期评估实验室及其模型，实验室可支付评估费用，前提是不影响评估方的独立性。Anthropic称，该法案体现了其Advanced AI Framework中的提案，公司自6月起一直向联邦和州立法者倡议这些政策。
Heat over time
Not enough continuous observations to draw a trend yet.