How moderation works

Strict rules at the door, human judgment where the consequence matters.

Checks apply across the product

Posts, comments, public names, Corner names and rules, links, flairs, profile text, ads, Context Notes and source URLs can all be reviewed. Uploaded media starts in quarantine. Safe files are approved automatically after type, size and ownership checks; flagged content is held for a closer review.

What is prioritised

Child safety, credible violence, non-consensual intimate material, exposed private information, active fraud and imminent real-world harm move ahead of ordinary quality or relevance disputes. A large number of reports does not automatically prove a violation.

Automation does not make final high-impact decisions by itself

English-language rules and risk signals help route work cheaply and consistently. They can be wrong. Account suspension, difficult context and appeals are reviewed by people where staffing permits. ReadChief does not currently claim that a paid generative-AI service reviews every contribution.

Community and platform authority

Corner moderators enforce their published local rules and can remove content, mute members or apply community bans. Platform moderators enforce site-wide rules, manage legal notices and can act across Corners. Every consequential moderation write enters an audit trail.

Notice and appeal

When it is safe and lawful, the affected user receives the policy category and action. They may submit a verified appeal. An independent reviewer should handle it where staffing permits. ReadChief may withhold details that would expose a reporter, facilitate evasion or create further harm.

Emergency and lawful disclosure

ReadChief may preserve or disclose limited records when required by a valid legal process or reasonably necessary to address an imminent threat, consistent with applicable law and the Privacy Policy.