Catfish Detector
Someone online looks a little too perfect, and something feels off. Upload one of their photos and find out whether it was generated by AI, lifted from another account, or heavily edited. This takes a minute and costs you nothing to find out.
What a fake profile photo looks like under analysis
Catfishing used to mean stolen photos of a real person. Since free face generators appeared, most fake profiles use faces that never existed — which is good news, because generated faces leave traces a stolen photo does not.
Eyes in the exact same place
Most face generators place the eyes at nearly identical coordinates in every image. Line up several of their photos: if the eyes sit at the same height and spacing every time, that is not coincidence.
A background that means nothing
Generated portraits have blurred, melted surroundings — no readable text, no coherent room, no recognisable place. Real photos have a real world behind the person.
Ears, teeth and jewellery
Asymmetric earrings, one ear a different shape, teeth merged into a single white block. Generators still fail on these consistently.
No camera in the metadata
A real phone photo carries the device, lens and capture settings. A generated or scraped image usually has none of that — or metadata from an editor instead of a camera.
Skin without texture
Pores, fine lines and sensor noise are missing. Our texture and noise analyzers measure this rather than relying on how it looks to you.
Hair and glasses that break the rules
Strands that end nowhere, glasses whose frames don't align, an earring that merges into hair. The eye skips over these; the analysis does not.
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.
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.
Signs worth checking against
- They avoid video calls — Always a reason. This is the single most reliable warning sign there is.
- The relationship moves very fast — Strong feelings within days, often before you have ever seen them live.
- Money comes up — An emergency, a customs fee, a can't-miss investment. This is the point of the whole exercise.
- Photos look like a photoshoot — Few casual pictures, no friends, no ordinary moments — just portraits.
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.
One thing this tool cannot do: if someone stole photos of a real person, forensic analysis may find nothing, because the photo is genuine — it simply is not them. For that case use a reverse image search alongside this check. The two together cover far more than either alone.
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.