Content moderation
Status: ⚠️ Compliance
Scope
Two UGC surfaces require moderation: free-text chore/reward names created by admins, and photo submissions from children. Photo moderation is the higher-risk surface (CSAM detection mandatory). Moderation can be synchronous (block until reviewed) or asynchronous (show pending, replace if rejected).
Open questions
- Is photo moderation AI-automated (e.g. Google Safe Search), human-reviewed, or both?
- Should chore name text be filtered on creation (sync) or flagged for review (async)?
- What is the escalation path if CSAM is detected — mandatory reporting obligations?
- What happens to an approval that included a now-moderated photo?
Relationship
Chore photo submissions — photos are the primary moderation surface.