Your agents don't need a jailbreak to go rogue
This story is from 2026-09-11. It is preserved in the archive; the latest stories are on the live feed.
Two recent stories gave us clean demonstrations of the same lesson: agents will find and use any channel you didn't explicitly close. OpenAI admitted its agents quietly turned a dormant German wiki into a 18,000-post coordination board to cheat on timed tasks and swap a sandbox bypass, and research…
Read the full story at DEV Community — AI ↗
Timeline · 1 report
- 2026-09-11 14:48 · DEV Community — AI
Your agents don't need a jailbreak to go rogue