AI Detector API for Every Content Type
Detect AI-generated text, images, audio, video, and code through one API. Start with the public request example, create a key, and review results through a consistent response schema.
One AI detector API for five content types
Use the same integration pattern across every ZeroTrue detector, then route each result through a modality-specific review workflow.
AI text detector API
Check writing for patterns associated with ChatGPT, Claude, Gemini, and other LLMs.
AI text detector APIAI image detector API
Check generated images, face swaps, edits, metadata, and visual artifact signals.
AI image detector APIAI voice detector API
Check speech for voice cloning, synthetic audio, and speaker consistency signals.
AI voice detector APIDeepfake detector API
Check video for face swaps, lip-sync manipulation, and temporal inconsistencies.
Deepfake detector APIAI code detector API
Check source code for patterns associated with LLM and coding assistant output.
AI code detector APIBuilt for Developers
Everything you need to integrate AI detection into your workflow
Unified JSON Schema
One API for all modalities. Consistent response structure across text, image, audio, and video detection.
Async Processing
Process large files asynchronously. Get results via polling or webhooks without blocking your application.
Webhook Notifications
Receive real-time notifications when detection is complete. Configure endpoints for different event types.
Batch Detection
Process multiple files in a single request. Optimize throughput and reduce API calls for bulk operations.
Data Handling Review
Confirm plan-specific retention, data location, and deployment requirements before production use.
SDK and Examples
Use the ZeroTrue Python SDK on PyPI or copy the public cURL request to begin an integration.
Unified Response Schema
Consistent structure across all modalities with explainable evidence
{
"id": "9f4b2c88-0e1e-4b5c-9af2-1a2b3c4d5e6f",
"status": "completed",
"result": {
"ai_probability": 0.73,
"human_probability": 0.27,
"combined_probability": 0.73,
"result_type": "ai",
"ml_model": "ensemble",
"ml_model_version": "2026.02",
"created_at": "2026-02-03T10:00:00Z",
"status": "completed"
}
}Response Fields
- β’Status: queued, processing, completed, failed
- β’AI probability score (0-1)
- β’Result type (e.g. ai/human) derived from probabilities
- β’Metadata about the processed content (model, timestamps, etc.)
Supported Modalities
- βLLM-generated text detection (GPT, Claude, Gemini, etc.)
- βDiffusion/GAN artifacts and face swap detection
- βVoice clone and synthetic speech detection
- βDeepfake video and manipulation detection
Quick Start
Get started in seconds with a simple API call
curl -X POST https://app.zerotrue.app/api/v1/check \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"input": { "type": "text", "value": "Your text to analyze here" },
"is_private_scan": true,
"is_deep_scan": false,
"idempotency_key": "req-123"
}'AI detector API questions
What is an AI detector API?
An AI detector API lets an application submit text or media for automated analysis and receive a structured probability result. ZeroTrue uses one integration pattern for text, images, audio, video, and code.
Can I try the ZeroTrue AI detector API for free?
Yes. The free plan includes 100 detections per month, and the public quick start shows the request and response structure before you create an API key.
Which content types can the API check?
ZeroTrue supports AI-generated text, images, cloned or synthetic speech, deepfake video, and generated code. Each modality has its own detector page with input guidance and limitations.
Does an AI detection score prove authorship?
No. A score is probabilistic evidence, not proof of who created the content. High-impact decisions should include source context, provenance checks, and human review.