video1.0<iframe src="https://www.loom.com/embed/eca3e42667d64175be79d0f7f87fd4ec" frameborder="0" width="1910" height="1432" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>14321910Loomhttps://www.loom.com14321910https://cdn.loom.com/sessions/thumbnails/eca3e42667d64175be79d0f7f87fd4ec-242edbb6ae8e2895-full.jpg995.072This Loom demonstrates how AI moderation settings control whether messages, nicknames, and images are hidden or only flagged. It explains that you can activate different moderation models, choose automatic hiding versus flagging, and view outcomes in an example chat where inappropriate text becomes hidden with toxicity reasons or highlighted without being hidden. It also covers automated user bans based on multiple users reporting the same message for the same user within a period, with optional admin confirmation for community managers. Finally, it recommends using pre moderation for images and avatars to prevent unverified content from appearing, and shows how nickname moderation resets offensive nicknames to a default for other users.