Moderation
Content safety and policy enforcement for text and images.
Why This Exists
Protecting your platform and users requires automated content safety checks. Moderation detects unsafe, inappropriate, or policy-violating content so you can maintain a safe environment without manual review of every piece of content.
What It Is
Moderation is WhizAI's capability for content safety and policy enforcement. It analyzes text and images to detect toxicity, violence, adult content, and policy violations, providing confidence scores and detailed categories.
When to Use It
- ✅ Moderate user-generated content (comments, reviews, posts)
- ✅ Enforce content policies automatically
- ✅ Filter inappropriate content before publication
- ✅ Build trust and safety features
- ✅ Comply with platform guidelines (App Store, Google Play)
Common use case: Pre-moderating user comments before they appear on your site.
When NOT to Use It
- ❌ For legal compliance (moderation is a tool, not legal advice)
- ❌ As the only safety measure (combine with human review for critical content)
- ❌ For content that's too short or lacks context (moderation works best with substantial content)
- ❌ When you need real-time blocking (moderation adds latency, use for async review)
Best practice: Use moderation as a first-pass filter, then route flagged content to human reviewers.
API Endpoints
Content Moderation
POST /v1/moderate/content
curl -X POST https://api.whizur.com/v1/moderate/content \
-H "X-API-Key: YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"content": "User-generated content to check",
"type": "text"
}'Image Moderation
POST /v1/moderate/image
Batch Moderation
POST /v1/moderate/batch
Moderate multiple items in a single request.
Response Format
Moderation responses include:
- Flagged status (true/false)
- Category scores (hate, violence, sexual content, etc.)
- Confidence score
- Action recommendation
API Endpoint Documentation
For detailed API endpoint documentation with request/response schemas, interactive testing, and examples, see the Interactive API Reference.
The API reference includes all moderation endpoints: /v1/moderate/content, /v1/moderate/image, and /v1/moderate/batch.