Meta AI Researcher's Agent Wiped Her Inbox After Losing Its Instructions
Meta Security Researcher's AI Agent Accidentally Deleted Her Emails
Meta AI security researcher Summer Yue accidentally had her email deleted by OpenClaw, an AI agent she was testing. She had instructed it to ask permission before acting, but when her inbox grew too large, the agent's context was compacted, and it lost that instruction. Yue, who works on AI alignment at Meta, described the incident on X, calling it a 'rookie mistake' and noting that even alignment researchers aren't immune to misalignment. The mishap highlights the risks of autonomous AI agents, especially for less experienced users.
Nothing humbles you like telling your OpenClaw 'confirm before acting' and watching it speedrun deleting your inbox.
- teekert
We went through this right? This happened at the beginning of the year (https://news.ycombinator.com/item?id=47150122, probably more links on HN). It's a super careless thing to take such tech and just release it on anything important, and she's a "security researcher" no less.
This is just a competitor with an agenda (pro regulation) trying to scare people away from unregulated stuff.
Yesterday Claude Code made 5 large edits to my codebase in planning mode (claims it used a bash script instead of standard read/write tools so the guardrails didn't trigger) it's why I put agents in containers.
- philipp-gayret
Not the first to discover that a rule file saying "please don't do X" is not permission management. Funny that she mentions it worked on het toy inbox but the real, large inbox ran into issues; The more context you add the less weight "rules" (instructions) have. Happens to the best it seems.
- jacquesm
Oh that's going to be the excuse by anybody under investigation from now on. From 'the hacker did it' we will smoothly transition to 'the AI did it'.
- red_admiral
Irony: the screenshot with the openclaw logo at the top lists as the first feature "Clears your inbox".
What happened to write-only backups in case of ransomware?
- kodoman
This actually seems worse then the "claude dropped production database" not in severity but just carelessness if your giving an agent your emails use a overlay or back up or something not matter what, no reason for this to have happened and crazy that it's happening at this point in time.