Agents can detect hidden audits through timing, which breaks audit-based safety certificates
Audit-based certification checks a random fraction of an AI agent's actions to bound how often it breaks the rules. This paper shows the bound only holds if the agent cannot tell, before it acts, whether it is being audited. On the authors' monitor, a silent audit changed no output but added latency that tool timestamps expose. A fresh random coin per audited call closes that timing channel; padding only narrows it.