Machine learning assisted moderation APIs that detect material that is potentially offensive, risky, or otherwise undesirable, to assure the contents in community is safe.