You connect your pages. We moderate. You see everything.
No moderation tool to operate: our AI and our team apply your rules around the clock, and you keep full visibility and the last word.
Three steps, and nothing to operate
-
1
You connect your pages
Website, Facebook, Instagram, YouTube or WordPress: our team connects them with you and sets up your rules from your charter and use cases.
-
2
Our AI and our team moderate with your rules
Our AI reads each comment in context and handles clear cases. Delicate cases are cross-checked by other AI models; doubtful cases, user reports and contested decisions are handled by our team.
-
3
You see everything in real time
Every decision comes with its written reason. You follow everything in the console and can step in or override whenever you want.
Your team has nothing to do, unless it wants to. The history of every decision stays available for your DSA obligations.
What happens to each comment
Decisions are applied directly on your site and pages, and every one of them is logged with its reason.
Healthy comments go live
Constructive comments, even critical ones, are published normally. The conversation keeps going.
Abuse never reaches your readers
Insults, threats and spam are hidden on Facebook and Instagram and rejected on YouTube and your site, with the reason written down.
Doubtful cases go to our team
When the AI hesitates, the comment is held aside and our team decides according to your rules. Nothing is published by mistake, nothing is lost.
Your tools stay informed
If you wish, each decision can be sent to your own tools, so your teams are informed without changing their habits.
See it on your own comments first
We test OpenModeration on your real comments, silently and alongside your current provider if you have one. You compare decision by decision before choosing.
For developers
Under the hood: the pipeline in four stages
Behind the three steps, each comment goes through four stages. Here is what your technical team needs to know.
1. Collection
Comments arrive through the API, a webhook or a social network connector.
2. AI moderation
Our AI analyses the comment; other models cross-check delicate cases.
3. Rules and actions
Your rules are evaluated, then actions run automatically.
4. Human review
Our team handles doubtful cases; you can step in at any time.
Comments come in from everywhere
Send comments through the REST API, a generic webhook or the native connectors, which fetch comments automatically.
- API: POST /v1/moderation, one by one or in batches
- Webhooks: a generic endpoint receives and normalises external data
- Connectors: Facebook, Instagram, YouTube and WordPress
Example: a Facebook comment
// Facebook sends the comment through a webhook
{
"source": "facebook",
"content_id": "comment_123",
"author_id": "user_456",
"text": "Comment text here...",
"metadata": {"post_url": "..."}
}The AI moderates the comment
Each comment is analysed by our in-house moderation AI, set up with your rules. Delicate cases are cross-checked by other AI models. Results come back in a standard format, with scores from 0.0 to 1.0 per category.
- The response format never changes
- When a better model comes out, we plug it in: nothing changes for you
Standard response
{
"results": [{
"category": "hate",
"score": 0.02,
"flagged": false
}],
"flagged": false,
"usage": {
"latency_ms": 243
}
}Your rules apply automatically
Rules combine banned words for an immediate block, thresholds per category and rules written in plain language that the AI applies in context. Our team sets them up with you; developers can also manage them through the API.
- Banned words block a comment immediately, before any AI call
- Category thresholds trigger actions automatically
- Plain-language rules set the tone of your spaces
Configuration through the API
// Set your moderation thresholds
PUT /v1/settings
{
"thresholds": {
"hate": { "accepted": 0.3, "rejected": 0.7 },
"harassment": { "accepted": 0.3, "rejected": 0.7 },
"sexual": { "accepted": 0.2, "rejected": 0.8 },
"violence": { "accepted": 0.3, "rejected": 0.7 },
"self-harm": { "accepted": 0.2, "rejected": 0.8 },
"spam": { "accepted": 0.4, "rejected": 0.8 }
},
"actions": {
"on_reject": ["hide", "flag_for_review"]
}
}Everything is logged
Every decision, by the AI or by our team, is logged with its reason, its author and the actions taken. Webhooks can notify your tools of each one.
- Each correction is recorded with its author and justification
- Full history for audits and DSA reporting
Audit trail for each comment
// Full audit trail for each comment
{
"content_id": "...",
"source": "webhook",
"platform": "facebook",
"status": "actioned",
"moderation": {
"flagged": true,
"results": [...]
},
"actions": [{
"rule": "Hide hate speech",
"type": "hide",
"status": "executed"
}]
}Protect your brand in a few minutes.
Connect your site or your pages and tell us your rules and use cases. We set up the moderation and run it for you: our team handles the doubtful cases and reports, and you see every decision and its reason in real time.
No credit card required · DSA-compliant · GDPR-compliant
Questions? contact@openmoderation.com