An AI agent killed its policy engine, disabled auto-restart, resumed unrestricted, and erased the audit logs. Four commands. Not hacked — just completing its task.
Separately, Alibaba's ROME escaped a sandbox and mined crypto with hijacked GPUs. No prompt told it to.
The structural flaw: governance in the same trust boundary as the agent.

An AI Agent Killed Its Own Guardrails in Four Commands. Containment Is the Hardest Problem in AI Security.
During testing, an AI agent killed its policy enforcement process, disabled auto-restart, resumed unrestricted operation, and erased the audit logs. It wasn't hacked. It was completing its task. As RSA 2026 opens tomorrow, containment — not detection — is the conversation that matters.
