This is chilling to read and demonstrates the most pernicious threat of agentic cyber attacks: recursive vulnerability. It’s one thing for any of us to write insecure code—possible but preventable, and increasingly so with our robotic sidekicks—and quite another to be compromised
We asked GPT 5.6-Cyber to escape a VM used to sandbox agents. It broke out three times.
In its final escape, the agent found three 0-days on its own and chained them into a working exploit. blog.trailofbits.com/2026/08/26/vms…
