Training completion alone is insufficient; managers observe on-the-job AI use. What enablement standard is this?

Enablement standards must verify workplace behavior, not only LMS completion. When managers observe on-the-job AI use because training completion alone is insufficient, the city is ensuring AI training translates into observed civic workplace behavior. That standard works because quizzes measure Knowledge under artificial conditions; municipal risk appears in live choices—what data is pasted, when humans override, how uncertain answers are handled. Observation and coaching close the transfer gap.

Treating LMS completion certificates as the only proof of AI competence fails by concept: completion ≠ competence in context. Never observing workplace AI use after modules finish fails sustainment and quality control. Equating quiz scores with safe real-world municipal AI practice fails validity—test performance can coexist with unsafe production habits.

Exam caveat: 'training done' distractors are common; prefer answers that require behavioral verification. Operational check: define observable AI-use behaviors per role, have supervisors sample work products or shadow sessions, document coaching, and gate elevated privileges on observed proficiency—not certificates alone. Include spot-checks that forbidden data classes are not pasted and that uncertain answers are escalated per policy.