Deepfake Detection API
A deepfake detection API is a REST endpoint that accepts an image or video and returns a structured verdict (AI-generated, likely AI-generated, likely authentic, or no AI detected) with a confidence score and an evidence chain. The FauxLens API exposes the same detection pipeline that powers our web tool, so content platforms, trust-and-safety teams, HR systems and dating apps can automate this at scale without building a detection stack in-house. Self-serve bearer keys, JSON responses, uploads deleted after analysis.
GET AN API KEYHow every scan runs
Read the methodProvenance
C2PA content credentials and EXIF metadata, checked first
Watermark & Pixels
OpenAI and Google AI watermarks read, then a detection model scores the image
Evidence Chain
Every finding itemised and graded, not reduced to a yes or no
- Files deleted after analysis
- No account required
What the Deepfake Detection API Does
The FauxLens deepfake detection API accepts an image or video file (or a URL to one) via a single authenticated POST request and returns a JSON verdict. Under the hood it reads C2PA content credentials and EXIF/XMP metadata, checks for the invisible watermark OpenAI and Google embed in their AI images (SynthID), has a specialist third-party detection model score the pixels, and falls back to a multimodal vision model, then returns a single verdict with a confidence score and an evidence chain.
This is the programmatic equivalent of the FauxLens web deepfake detector. The web tool is for one-off human checks; the API is for automating deepfake detection inside your own product - screening profile photos at signup, scanning uploaded media before publication, flagging suspected synthetic media in submitted video, or triaging user reports at volume. Every call returns the same structured evidence a human analyst would see. So you can set your own confidence thresholds and route flagged content into review.
Endpoint, Authentication, and Rate Limits
Authentication uses an API key bearer token passed in the Authorization header. Developer keys are limited to 20 requests per minute; there is no monthly request cap, because your credit balance is the quota. Enterprise volumes raise the per-minute limit and can additionally restrict a credential to specific IP ranges. Rate-limited responses include a Retry-After header.
Read moreShow less
Supported inputs: JPEG, PNG, WebP, HEIC and AVIF images; MP4, MOV, WebM and MKV video. A developer key is capped at 25MB per request and a URL fetch at 50MB; enterprise credentials get 200MB. Both multipart form-data uploads and a JSON body with a url field are supported. All traffic is encrypted in transit, and submitted files are deleted once the verdict is returned, the same as on the web tool. Worked curl and Python examples are on the developer API page; there are no SDKs to install, because the API is one POST.
Sample JSON Response
Every deepfake detection API call returns a consistent JSON envelope: a summary verdict (one of "AI-Generated", "Likely AI-Generated", "Likely Authentic" or "No AI Detected"), a confidence_score between 0 and 1 that is not a calibrated probability, an is_short_circuited flag for scans that ended early on a decisive check, a short_id, and an evidence_chain array. Each entry has a layer (the check that produced it), a status (passed, warning, flagged, or info), a human-readable label, and a plain-language detail. There are no per-signal numeric scores or generator attribution. Your integration can act on the summary for a simple allow/deny gate, or show the evidence chain in a review UI for your trust-and-safety team.
Who Uses a Deepfake Detection API
Dating and social platforms use the API to screen profile photos and uploaded media at signup and on report, catching AI-generated personas used in romance scams before they reach other users. Trust and safety tools scan submitted video and profile images to flag suspected synthetic media, a documented and growing fraud vector. Newsrooms and content platforms scan submitted imagery before publication. Insurance and claims systems triage submitted damage photos. Marketplaces verify listing images. In each case the value is the same: deepfake detection that runs automatically on every piece of content you send instead of relying on a human to notice.
Ready to verify an image?
Get an API KeyFrequently asked questions
One call costs 10 credits, the same as one scan in the web app. Credit packs are $2.99 for 500 credits, $9.99 for 2,000 credits, $19.99 for 5,000 credits. Your free starting credits cover a few calls, once. There is no monthly free allowance and no subscription: credit packs are a one-time purchase and do not expire. Enterprise adds prepaid volumes (invoiced) and higher rate limits.
They run the same detection pipeline and return the same evidence. The web tool is a browser interface for one-off human checks; the API is for automating detection inside your own application (screening media at scale, on upload, or on report) and returns machine-readable JSON instead of a rendered report.
No. Submitted files are deleted once the verdict is returned, exactly as on the web tool, and are never used to train a model. The response carries a short_id that identifies the scan: what can be kept is the result, not your file.
It depends on the file and on which checks run. A scan that ends early on a content credential returns sooner than one that goes through the full pipeline, and video takes longer than a still image because frames are extracted and checked. Responses have no timing field, so measure latency on your side.
Create an API key, then send POST /v1/analyze on api.fauxlens.com with the key as a bearer token. Send the file as multipart/form-data, or a JSON body with a url field. The response carries the verdict, a confidence score and the evidence behind it.
No SDK is needed. The API is one HTTPS POST, so any language that can send a form upload or JSON can call it. A worked curl example is on the developer API page.
It checks a submitted selfie or ID photo for AI generation and AI edits and returns the evidence. It is not a liveness check and does not match a face to a document, so use it alongside your KYC provider.
Keep exploring
Guides
- AI Image Detection API: A Developer's Integration GuideBuilding a platform that needs to verify image authenticity? This technical guide covers API integration for AI image detection, from authentication to response handling to production deployment.
- The Science of Deception: How AI Detection WorksA detailed guide to the digital forensics of AI image detection, from Error Level Analysis (ELA) and JPEG compression artifacts to Photo Response Non-Uniformity (PRNU), and which checks a FauxLens scan actually runs.
- Sora, Veo 3, and Kling: How to Detect AI-Generated Video in 2026AI video generation has crossed a threshold. Sora, Veo 3 and Kling now produce clips the untrained eye cannot separate from real footage. Here is what forensic analysis can still find, why a re-encoded clip can look more authentic than the original, and where video detection genuinely fails.
More tools
Detect while browsing: try the Chrome extension
Right-click any image · No account required