AI safety mechanism causing claim suppression when confidence is low. Triggers: single-source claims, ambiguous entities, poorly structured assertions, claims lacking EAV-E compliance. Manifests as hedging (“some companies claim”), generalization (category-level response instead of entity-specific), or omission. Overcome through corroboration (5+ tier-1/2 sources), entity disambiguation (KGMID), and EAV-E compliant content. Test: ask AI specific claim about your entity, observe whether it’s cited or hedged. Hedging = insufficient structural evidence.