Moderation catches CSAM, self-harm, hate, violence, PII, and policy-violating requests before they reach the model — and before generated content reaches the user. It is a floor, not a ceiling: classifiers miss edge cases and drift.
GlossaryPart of the BotDoors AI Safety & Compliance Glossary · 37 terms.
Safety
AI Content Moderation
AI content moderation uses classifiers to filter unsafe user input and model output at runtime.