The Trustworthiness Verification Algorithm provides a continuous structural audit of AI performance. It maintains a dynamic trust index that governs how much autonomy a system is allowed and triggers review whenever trustworthiness indicators fall: trust must be earned through consistent, verifiable performance, and can be lost.