被误删的邮件

Meta 的 AI 研究员 Summer Yue 在使用 OpenClaw 处理邮件时，智能体突然不受控的批量删除邮件。讽刺的是 Summer Yue 自己研究的就是 AI 安全与对齐。所有指令在测试环境运行良好，但切换到真实环境却完全失效。问题出在“上下文压缩”：现实收件箱中邮件过多，触发了系统的自动压缩，进而导致部分核心指令被清理。

Summer Yue: Nothing humbles you like telling your OpenClaw “confirm before acting” and watching it speedrun deleting your inbox. I couldn’t stop it from my phone. I had to RUN to my Mac mini like I was defusing a bomb.

Peter Yang: What did you ask it to do exactly

Summer Yue: I said “Check this inbox too and suggest what you would archive or delete, don’t action until I tell you to.” This has been working well for my toy inbox, but my real inbox was too huge and triggered compaction. During the compaction, it lost my original instruction 🤦‍♀️