Wu Shixiong's Large Model Academy

Aug 27, 2026 · Artificial Intelligence

Why Agents Refund When Policy Says No: The Retrieval-Enforcement Gap

An AI agent correctly retrieves a 'no refund' policy but still executes a $50 refund, exposing the critical gap between policy retrieval and runtime enforcement; the article argues authorization must be enforced at the tool gateway with bound decisions, not just in model context, and outlines regression tests for deny paths.