Event ·Multi-source

Anthropic 致信加州州长纽森:建议法院有权阻止高风险 AI 模型发布

据路透社 10 月 29 日报道,Anthropic 于 10 月 15 日致信加州州长纽森,建议政府可向法院申请禁令,阻止对公众构成严重风险的 AI 模型发布。Anthropic 反对设置关闭整个模型的紧急关闭开关,认为这会让官员产生虚假安全感,更倾向关闭出问题的智能体或暂停特定客户权限。纽森 9 月召集 AI 专家小组制定安全建议,OpenAI 也于 10 月 20 日提交了意见。此前 6 月,美国政府曾要求 Anthropic 限制外国用户访问 Mythos 5 和 Fable 5 模型,商务部随后撤销该命令。

Current read

Anthropic 给加州州长写信谈 AI 监管,反对一刀切的关闭开关,主张法院禁令这种更精细的手段,还提到了自家被限流的往事。

11 sources57° AI heatLast updated 2026/10/28 23:05:14

Evidence

11 sources
Related source 1Andrew Curran
View source
Related source 2Detik Inet
View source
Related source 3kimmonismus
View source
Related source 4Tech in Asia
View source
Related source 5Yuchen Jin
View source
Related source 7CSET Georgetown
View source
Related source 8rohanpaul_ai
View source
Related source 9爱范儿
View source

Conflict check

Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.