Fake listing photo and seller document detection

Yes. Listing images are checked for generator traces such as repeated textures, warped text and impossible details, for edited regions, and for missing or conflicting camera data. Each finding is marked on the image with its strength, so a reviewer can decide whether to ask the seller for originals before the listing goes live.

It checks the files sellers submit: ID images, business registrations, bank details and proof of address are checked for editing, AI generation and rebuilt templates. Trustable does not perform identity database lookups or liveness checks; it adds a forgery check to the onboarding you already run. See onboarding and KYC.

Yes. Verification and listing video is checked for face swaps and other generation traces, and findings are marked on the frames where they appear. As with photos, a finding is a reason to look closer or ask for a live check, not a decision on its own.

Through the API or the Trustable web app. Send an image, video, document or ID at upload or when a report comes in, and get back the findings, where they are and how strong each one is. Routing results into your moderation queue is in development.

No. A flag means a file needs a closer look. Product photos are edited for innocent reasons, such as background removal or colour correction, and that can look like manipulation. Trustable shows what it found and how strong the signal is; your team decides what happens to the listing or the account.