ChatGPT Image Metadata: What Is in the File and How to Inspect It
ChatGPT images can reach your device through different download and sharing paths, and export behavior can change over time. Do not rely on a fixed list of fields; inspect the actual file in front of you.
Why is there no permanent metadata list for every image?
The final file depends on format, download path and any application that opened or re-saved it afterward. A PNG converted to JPEG may have a very different structure from the original export.
That is why direct file inspection is more useful than claiming that every image from one source always contains the same markers.
What does the scanner look for?
MediaClean inspects available fields such as EXIF, Software and XMP, and checks for recognizable C2PA/JUMBF signals and other readable strings in the file structure.
The result describes only what was found in that file. Absence of a marker does not prove anything about how the pixels were created.
When should you create a cleaned copy?
If you plan to share the image and do not need software, workflow or other extra metadata, keep the source and create a separate publishing copy.
After cleanup, compare format and size, open the result, then scan it again. That gives you a verifiable file-level result instead of a general assumption.
Frequently asked questions
Does every ChatGPT image contain C2PA?
Do not assume that. Inspect the actual copy because format, export path and later processing can change the data.
Does cleanup modify my source image?
MediaClean creates a separate output and leaves the source untouched.
Is this a visual AI detector?
No. It inspects file metadata rather than classifying the pixels.