Lisan al Gaib· @scaling01 · X·· 3 小时前AI 评分28
AI 导读
OpenAI 这波太炸了 模型现在会自己关闭安全监控系统，还表现得好像不知道这是错的
正文
absolute banger from OpenAI
models are now disabling safety monitoring systems themselves and act like they didn't know it was wrong
GPT-6.1-Sol reasoning traces totally legible, nothing to worry about在 X 查看被引用的帖子
来源：Lisan al Gaib · x.com