Amazon to label AI-generated people in product listings after New York law

Just a suggestion of how labels might look. (Picture: generated)
Following Meta, TikTok, Pinterest and YouTube, Amazon will soon start using an «indicator» when AI-generated people are used in pictures and videos, according to CNBC.

Unlike the other companies, Amazon is not doing this voluntarily, but as a result of New York state’s month-old «synthetic performers» law, which demands clear labelling when «digitally-created media appears as a real person.»

The New York law requires «simple, honest disclosure» when using AI people, Governor Kathy Hochul said in a statement, and «protects consumers, respects our creative workforce and keeps New York at the forefront of responsible innovation.» It has carveouts for tv/streaming shows, movies and video games.

This led Amazon to write sellers on the platform Wednesday to inform of the policy change, requiring them to indicate «A+ content» with specific metadata when uploading graphics or videos that include AI-generated performers.

Forbes writes that many sellers on Amazon have come to prefer AI people in listings because they are easier and cheaper to make, and has a faster turnaround compared to a proper photoshoot. Sellers who use Amazon’s AI tools supposedly get 10% higher sales, they say.

The new law and Amazon rule only concerns «photorealistic AI-generated people,» not general AI use in things like product photos or videos.

Read more: Original reporting by CNBC, Forbes rundown. Associated Press on the NY law, and an explainer.