Moderation & Appeals Policy
Layered moderation uses product limits, automated checks, accountable owners, reports, admin escalation, and simple appeals.
Moderation layers
The moderation model is layered: product limits first, automated safety checks second, accountable owner moderation third, user reports fourth, and admin/steward escalation for serious or disputed cases.
AI or automation may flag, hide, or suggest action later, but serious enforcement decisions and appeals require human responsibility/review.
Report reasons
Users should be able to report spam/scam, illegal content, four-principle violations, harassment, gossip/accusations, health danger, legal/financial danger, child safety concerns, impersonation, privacy/doxxing, wrong category, or other issues.
Reports should be handled through queues rather than manual feed monitoring so the platform remains operable by a small team.
Possible actions and appeals
Moderation actions may include no action, request edit, hide temporarily, remove, lock Q&A, restrict visibility, warn, suspend posting, suspend account, mark scam/fraud, escalate, restore, or approve after review.
A simple appeal/review path should exist through a request review button, report/contact page, or support email.