Jan Leike：Claude 4 能力太强，Anthropic 按负责任扩展政策启用额外安全缓解措施
So many things to love about Claude 4! My favorite is that the model is so strong that we had to turn on additional safety mitigations according to Anthropic's
AI 导读
Anthropic 的 Jan Leike 表示，Claude 4 让他最喜欢的一点是模型能力足够强，以至于团队不得不按照 Anthropic 的负责任扩展政策（responsible scaling policy）开启额外的安全缓解措施。
推荐理由
AI 评分 52
Anthropic 对齐团队负责人以内部视角说明 Claude 4 能力提升触发了负责任扩展政策下的额外安全缓解措施。
主题
正文
当前显示原文，中文全文译文尚未就绪。
So many things to love about Claude 4! My favorite is that the model is so strong that we had to turn on additional safety mitigations according to Anthropic's responsible scaling policy