Add the image
Drop, paste or choose a picture, or take a photo on your phone. Use the original file if you can.
Check the file’s Content Credentials and AI metadata, then get an AI-generated likelihood from a detector that runs on your device. Nothing is uploaded, and there is no limit on checks.
Open AI Image Detector →Drop, paste or choose a picture, or take a photo on your phone. Use the original file if you can.
If the file carries Content Credentials or AI metadata, you see what it says right away.
The detector gives an AI-generated likelihood, a verdict and how confident it is.
Use Manlin to reflect, plan and take your next step.

Yes. Check as many images as you like, with no account and no limits.
No. Both checks run in your browser and the image never leaves your device. The detector is downloaded once and kept in your browser.
It gets most clear-cut cases right, but it is an estimate and can be wrong. Treat a low or medium confidence result as uncertain, and never use it as the only proof that someone used AI.
Content Credentials (C2PA) are signed notes inside a file that say how it was made. Several AI image tools add them, as well as an IPTC tag that marks AI-generated media. This tool reads what those notes say; it does not verify the signature.
Most social networks, messaging apps and screenshots remove metadata, so its absence tells you nothing about whether AI was used.
JPG, PNG, WebP and AVIF in any modern browser, and HEIC photos where the browser can open them (Safari). Files up to 50 MB.
Only when the file’s metadata names the generator. The estimate itself only says how likely the picture is to be AI-generated.