Custom moderation that fits your community.
Rewrite, redact, escalate, or refuse — per category, per surface, per policy. Not one-size-fits-all toggles.
UGC platforms and marketplaces need moderation that matches their community standards, not a generic provider preset. Policy Gateway lets your trust & safety team author and version policies as code, A/B test outcomes, and stream every decision into your moderation queue.
Why teams in trust & safety hit a wall.
Generic moderation APIs miss your context
A dating app, a developer forum, and a marketplace listing all need different rules. Provider-side moderation flattens that nuance into a handful of toggles.
Block-or-allow is too crude
Real moderation often means rewriting (sanitize), redacting (strip PII), or escalating (human review). Off-the-shelf APIs only label.
No way to audit individual decisions
Regulators (DSA, KOSA) increasingly require justification for every moderation action. Most APIs return scores, not reasons.
Built for trust & safety workloads.
Policy-as-code per surface
Define one policy for marketplace listings, another for direct messages, another for public posts. Version, simulate, and roll out per-policy with shadow and canary modes.
Outcomes beyond block
Configure rewrite (sanitize), redact (strip PII), escalate (human queue), or refuse — per category, per surface. Tune outcomes without redeploying your app.
Decision-by-decision audit
Every moderation action logged with policy ID, reason code, and content hash. Stream into your moderation tooling, T&S queue, or compliance archive.
Scenarios from the field.
Marketplace listing sanitization
Auto-rewrite listings that violate community guidelines (prohibited terms, bad formatting) instead of bouncing the seller. Conversion stays up; policy stays enforced.
DM PII redaction
Strip phone numbers, emails, and government IDs from user-to-user messages before delivery. Reduces off-platform fraud routes without breaking conversations.
High-risk escalation
Route specific content categories (self-harm, threats, CSAM signals) to your trust & safety queue with full conversation context attached for fast triage.
Designed for the frameworks your auditors care about.
Built around the regulatory frameworks reshaping platform moderation in 2024 and beyond.
- DSA Article 16/17Decision metadata supports notice-and-action and statement-of-reasons obligations.
- GDPR-alignedPII redaction at the API boundary; configurable retention.
- COPPA-aware toolingPer-surface policies for products with under-13 users.
- Shadow + canary modesRoll out new rules without surprising your community.
- Per-project key scopingIsolate keys per surface, app, or product line.
- SOC 2 (in progress)Enterprise audits underway.
Ready to bring governance to your trust & safety stack?
Talk to an engineer about your deployment, or grab an API key and start building today.