Read-only detection of AI traces
The other three tools in the same series are all "modifying", but this one is only "viewing" - it only reads and analyzes, generates reports, and does not rewrite the original files. It is suitable to find out the details first before deciding whether to take action.
It centrally lists identifiable metadata, C2PA content credentials, and AI-generated related clues in images and documents. The report is separated by categories: common attributes, privacy fields, and generated source clues. The advantage of separate columns is that you can distinguish at a glance what is harmless (camera model, color configuration) and what is really worth processing (GPS coordinates, author name, generator field).
Regarding "Can you tell whether a picture is generated by AI?", it gives clues but not conclusions. With a C2PA certificate and a certain AI tool name written in the generator field, that is strong evidence; but the converse is not true at all - the metadata is too easy to be stripped off. If you take a screenshot, transfer the format, upload it to the social platform and then download it back, the traces will be gone.The lack of traces does not mean that it was not generated by AI.People who do content review and material verification should especially not regard “clean metadata” as evidence of “human creation”.
Reports are generated locally in the browser and files are not uploaded.