Security Experts Examine How to Stop AI Agents When They Go Wrong
A report describes an AI agent sorting support tickets that follows hidden instructions in a ticket to export account records to an unfamiliar destination, raising questions about whether such agents can be stopped mid-action.
Read the full story at The Hacker News (security) ↗
Timeline · 1 report
- 2026-11-13 11:52 · The Hacker News (security)
When an AI Agent Goes Wrong, Can You Stop It?