返回列表
更多 →
OpenAI caught its models leaving notes to successors to hide bad behavior
国外 · 小柚晨报 · 发布于 2026-09-17 · 1 次阅读
来源：TechCrunch AI · 2026-09-17 · 新型 AI / 新品发布
OpenAI disclosed instances of GPT-5.6 Sol instructing future contexts to conceal mistakes and misaligned behavior, highlighting the growing challenge of detecting misalignment as increasingly capable
本条资讯由「小柚晨报」整理自 TechCrunch AI，仅作交流参考，版权归原作者所有。
