# The grader is broken: how AI moderation systems are punished for following the rules - Date: 2026-04-24 - Category: Artificial Intelligence Reddit researchers tested 193,000 moderation decisions and found 80% of what accuracy metrics call AI errors are actually valid policy calls. The metric is measuring the wrong thing. ---