Scam AI detection API

curl -X POST "https://api.scam.ai/v1/detections" \
  -H "x-api-key: <YOUR_API_KEY>" \
  -F "file=@/path/to/photo.jpg"

One endpoint. Send an image, a video, a recording or a video link with your API key. The server works out what you sent, runs the right detectors, and answers in one JSON body with a verdict, a score and the credits it charged.

1. Get a key

Create a key in the dashboard. It is shown once. Keep it on your server: a key grants full access to your account, so it never belongs in a browser, an app bundle or a public repository. Every request carries it in the x-api-key header.

x-api-key: <YOUR_API_KEY>

2. Send a file

Post the file as the file field. The official SDKs (@scam-ai/sdk on npm, scamai on PyPI) read SCAMAI_API_KEY from the environment and wrap the same call.

curl -X POST "https://api.scam.ai/v1/detections" \
  -H "x-api-key: <YOUR_API_KEY>" \
  -F "file=@/path/to/photo.jpg"

3. Read the verdict

{
  "id": "6f1c2a3e-9b7d-4e5f-8a21-3c4d5e6f7a8b",
  "status": "completed",
  "created_at": "2026-09-15T10:12:33.123Z",
  "media": {
    "type": "image",
    "filename": "photo.jpg",
    "mime_type": "image/jpeg",
    "bytes": 248113
  },
  "model": "Eva V1.6",
  "verdict": "LIKELY_AI",
  "score": 0.92,
  "summary": "This image shows strong signs of being AI-generated or edited.",
  "credits_used": 1
}

Branch on verdict. LIKELY_REAL means let it through, ALERT means put a human on it, LIKELY_AI means block or escalate. score is the probability the media is AI-made or manipulated, 0 to 1. Both are explained on Reading the result.

Next: Which input do I send? No key yet? Book a 15-minute demo to get access.