AI Image Detector API
The same engine behind the web app, as a REST endpoint. Send an image, get back a verdict, a confidence score and — unlike most detection APIs — the individual score from each of the twelve methods, so you can set your own thresholds instead of trusting ours.
What the API returns
Most detection APIs hand you a single number and expect you to trust it. Ours returns the working, because a moderation pipeline needs thresholds you control.
A verdict and a score
AI, REAL or UNCERTAIN, with a confidence percentage. UNCERTAIN is a real outcome rather than a rounding of a coin flip.
All twelve method scores
Each analyzer reports separately, so you can weight them yourself — raise the bar on metadata for user uploads, lower it where files are pre-processed.
Video frame analysis
Clips are decomposed into frames and scored across them, with temporal consistency included.
Predictable rate limits
Documented per plan, with remaining quota returned in response headers on every call — no guessing where you stand.
A test key that costs nothing
Separate test credentials for integration work, so development does not eat production quota.
Nothing retained
Images are analysed and discarded. Nothing is stored, and nothing feeds training data — which matters when you are processing your users' uploads.
How the analysis works
Upload the original file
Full resolution, straight from the source. Screenshots and re-saved copies destroy most of the evidence we measure.
12 independent methods run in parallel
Three neural networks look at the image as a whole. Nine forensic methods examine it as a signal: frequency (FFT) analysis, noise pattern, pixel-level statistics, EXIF metadata, Error Level Analysis, texture coherence, colour distribution, Benford's Law and edge detection.
You see the breakdown, not just a number
Every method reports separately, so you can tell whether the verdict rests on one weak signal or on nine that agree.
To try the same engine by hand before writing any code, Start with our AI image detector — it covers any image type and runs the same twelve methods.
How much each method counts
Most detectors give you one number from one model. We run twelve independent analyzers and weight them — three neural networks carry the majority, but nine forensic methods examine the file as a signal and can overrule them. Here is the exact weighting we use in production:
| Method | Weight in final score |
|---|---|
| Neural network #2 | 25% |
| Neural network #1 | 15% |
| Neural network #3 | 15% |
| EXIF metadata | 10% |
| Frequency (FFT) | 6% |
| Benford's Law | 6% |
| Noise pattern | 5% |
| Error Level Analysis | 4% |
| Texture coherence | 4% |
| Edge detection | 4% |
| Pixel statistics | 3% |
| Colour distribution | 3% |
What people actually upload
These are our own numbers, not an industry estimate — every check run through FakeSec since launch. The split is closer than most people expect: roughly four in ten uploads really do turn out to be generated.
What teams build with it
- Marketplace moderation — Flagging generated product imagery at intake.
- Dating platforms — Screening profile photos before they go live.
- KYC pipelines — Selfie verification alongside liveness checks.
- Newsrooms and archives — Bulk-checking incoming or historical material.
Accuracy — and where it breaks down
On our internal benchmark of 240,000 images (120,000 AI-generated, 120,000 real) the combined score reaches up to 98% on unedited, full-resolution files. That number drops, sometimes sharply, in four situations — and you should know them before trusting any result:
- Heavy compression. Social platforms re-encode everything. A photo pulled from Instagram has already lost much of its noise signature.
- Screenshots. A screenshot is a new image of an image. Original metadata and compression history are gone.
- Stylised content. Flat colour and clean lines remove the texture and noise evidence forensic methods depend on.
- Post-processing. Added grain, filters, upscaling and manual retouching all mask generation traces.
No detector on the market is 100% accurate — ours included. Treat a high score as strong evidence, not as proof. When a decision matters, look at which of the 12 methods agreed.
For pipeline design, one number matters more than accuracy: what you do with UNCERTAIN. Auto-rejecting on it will generate false accusations against real users. Route those to human review — that is what the flag is for.
Frequently asked questions
Related tools
Check your first image now
Create a free account and get 3 checks. Full breakdown across all 12 methods, no card required.