What we analyse today
Still images in common formats, including JPEG, PNG, WebP, and HEIC. We look for supported face swaps and supported image manipulation, and for signs of synthetic generation in images where no face is found. On a photo of a person, AI-generated detection is recorded but does not decide the result until it has been validated on photos of people.
Voice and video files are accepted for analysis, but they are in testing and return inconclusive. They do not produce production decisions.



