There is a line in enterprise AI that nobody draws carefully enough: the line between an agent that answers and an agent that acts.
A chatbot quoting a stale wiki page is embarrassing. An agent issuing a refund, rotating a credential, or deleting a queue because its plan said so is an incident with a dollar figure and a postmortem. The difference is not how smart the model is. It is whether anything stands between the model's plan and the real world.
Read the original article:
Deny by Default: Policy-as-Code Guardrails for Agents That Act
