Trust & safety
Built to be trusted
Safety and quality are not add-ons. They are built into the platform from the ground up.
AI moderation
Every submitted idea passes through AI moderation before it is published. AI checks for harmful content, spam, dangerous instructions, abusive language, and low-quality submissions. High-risk content is blocked; borderline content is escalated to human review.
Human governance
AI recommends — humans decide. Admins review all flagged content and make final moderation decisions. No permanent bans happen without human confirmation. High-risk decisions are logged and auditable.
Spam and low-value filtering
The platform is designed for serious ideas. AI and community voting together surface high-quality content and bury spam, empty submissions, and low-effort posts. Repeated violations result in account review.
Reporting
Anyone can flag content for review. Reports are reviewed by human admins. Urgent safety issues are prioritised. Reporters receive a response when the case is resolved.
Respectful debate
Ideas can be challenged and criticised — that is healthy. Personal attacks, harassment, and hate speech are not. Community guidelines define the line clearly.
Transparency and audit logs
All moderation decisions — AI and human — are logged. Admins can audit the full history of any decision. Users whose content is removed receive a reason.
For the full content policy, see the Community Guidelines. To understand AI moderation in detail, see AI Transparency. To report a safety issue, use the contact page.