Zero GPT

Deepfake Image Detector

Updated

This deepfake image detector looks for signs that a face in a photo has been manipulated, such as a face swap or edited features. Upload an image up to 15 MB to get a face-manipulation score, plus a separate score for whether the whole image looks AI-generated. Both are probabilities to weigh, not verdicts.

Limitations: Every result is a probability that can be wrong, not proof of how an image was made. How results can be wrong →

Open the Image Detector →

3 free scans in total, shared across all five tools (not per day or per month). No signup needed to start.

How the Deepfake Image Detector Works

The image is sent to a third-party image-analysis provider (named on the How It Works page) that runs two models: one for AI-generated images and one for face manipulation. The deepfake score comes from the face-manipulation model.

Because it looks at faces, the deepfake score is most meaningful on photos where a face is clearly visible. On landscapes, objects or text, only the AI-generation score is useful.

Deepfake vs AI-Generated: Two Different Questions

  • Face manipulation: a real photo where someone's face was swapped or altered.
  • AI-generated: an image created from scratch by a generator, with no original photo.
  • Both: a generated scene with a real person's face added; both scores may rise.
  • Neither: a real, unedited photo; both scores should be low, but can still be wrong.

Limits to Keep in Mind

Compression, resizing and screenshots remove detail, so images saved from social media are harder to judge than originals. Heavy retouching and beauty filters on real photos can raise the face-manipulation score.

This page covers still images. For video, use the AI video detector, which reports an AI-generation probability with ML, OCR and metadata sub-signals; it is not a dedicated face-swap detector.

What to Do With a High Deepfake Score

Look for the original source and an earlier copy of the image; an unaltered version elsewhere is the clearest evidence of manipulation. Check the face against other photos of the same person: lighting, skin texture and the edges around hair and jaw.

If the image could harm someone, be careful about sharing it further, even to debunk it. Report it to the platform where you found it.

Frequently asked questions