Canva AI Detector
Generation built into a design tool reaches a different audience: people making a presentation or a social post who do not consider themselves AI users at all. That is why generated imagery arrives undisclosed in marketing material far more often than through deliberate deception.
What design-platform output looks like
Images generated inside a design tool usually leave that tool as part of a composition, which changes what you are looking at.
The software field
Design platforms commonly leave their name in the exported file's metadata. Not proof of generation — the same field appears on hand-made designs — but a useful starting point in [[/exif-viewer|our EXIF viewer]].
Composites, not single images
Generated elements sit inside a layout with text and shapes. Only part of the image is synthetic, which makes it a manipulation question as much as a generation one.
Export flattening
Everything is flattened on export, giving the whole file one uniform compression history and hiding the region boundaries that would otherwise show.
Stock and generated mixed freely
A single design often combines licensed stock, generated elements and original photography — a mixture no single verdict describes well.
Undisclosed by default
The audience here is not hiding anything. They simply do not think of a built-in feature as “using AI”, which is why disclosure rarely happens.
Generation statistics still apply
Where a generated element is large enough, the usual signals hold: flat noise, even detail, no capture data.
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.
For images from outside a design tool, 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.
Where this matters
- Advertising compliance — Rules on synthetic imagery in commercial material.
- Agencies and clients — Deliverables specified as photography.
- Publishers — Editorial policies on generated illustration.
- Education — Coursework in design and marketing.
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.
Composites are the hard case here. If a generated element occupies a small part of a busy layout, a whole-image verdict may miss it — the manipulation methods that look for regional inconsistency are more useful than the generation score.
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.