Amazon policyEffective 22 July 2026All worldwide stores
Drop in the listing images, A+ images or product videos that show a photorealistic AI-generated person. This writes the keyword Amazon looks for into the XMP dc:subject field as an rdf:li inside the rdf:Bag — then reads the finished file back and shows you what is in it.
Everything happens in this tab. Your files are never uploaded and never leave your computer. Pixels and video streams are not re-encoded — only a metadata block is inserted, so the file you get back is the file you put in, plus the tag.
Drop files here, click to choose, or paste
One file or a whole batch. Filenames are kept exactly, so you can re-upload straight to Seller Central.
JPEG · PNG · MP4 · MOV | TIFF, GIF, WebP, WebM → use the exiftool command below
Why this is required
Amazon Seller Central · Policy and Compliance · effective 22 July 2026 · worldwide stores
If a listing image, lifestyle image, product video or A+ module shows a photorealistic person who was generated by AI, Amazon wants that disclosed in the file’s metadata before you upload it. The mechanism is an IPTC/XMP keyword: the value contains-synthetic-performer has to appear as an rdf:li inside the rdf:Bag of the XMP dc:subject field.
When the tag is there, Amazon shows shoppers an indicator saying the image contains AI-generated people. The rule follows disclosure legislation on advertising that features synthetic performers, and Amazon has applied it across all of its stores rather than only the market that prompted it.
Does it apply to this file?
Using an AI tool somewhere in your workflow is not what triggers this. What triggers it is a person in the frame who does not exist. That is a call only you can make — so the decision stays with you, and this page only writes what you decide.
There is a photorealistic AI-generated person in the image or video.
No tag is needed when the image or video:
Other ways to write the same keyword
Three formats cover almost every Amazon asset, and those are the ones handled here. For anything else — TIFF, GIF, WebP, WebM, AVIF — or for tagging a few hundred files in one go, exiftool writes the identical structure.
Install exiftool, then run this in the folder holding your files. The += adds the keyword without wiping keywords that are already there.
exiftool -XMP-dc:Subject+=contains-synthetic-performer -overwrite_original "your-file.webp"
A whole folder, recursively, images and videos alike:
exiftool -r -XMP-dc:Subject+=contains-synthetic-performer -overwrite_original .
Drop -overwrite_original if you would rather exiftool kept a _original backup of every file beside it.
Worth doing on one file before you trust a batch of three hundred. This prints the keywords in the XMP dc:subject field:
exiftool -XMP-dc:Subject -s "your-file.jpg"
To see the raw XMP packet exactly as it sits in the file, including the rdf:Bag structure Amazon reads:
exiftool -xmp -b "your-file.jpg"
Or just drop the file back into the box at the top of this page — it reads the tag out and shows you every keyword it finds.
macOS, Preview: open the image, then Tools ▸ Show Inspector ▸ the keywords tab. Add contains-synthetic-performer as a keyword and save. Preview writes it to XMP dc:subject.
Windows, File Explorer: right-click the file ▸ Properties ▸ Details ▸ Tags, type the keyword, then Apply. Explorer writes this for JPEG and TIFF; for PNG and video it does not, so use the box above or exiftool.
Either way, check the result before a bulk upload — some editors write to a different field that Amazon will not read.
Send us one ASIN and we will send back a real audit — three findings with the category numbers behind them, a content benchmark against the listings ranked above you, and main-image concepts built from your own photography. Free, one per brand.
