Following two explicit, AI-generated images of female public figures on its sites, Meta’s Oversight Board plans to look into how Facebook and Instagram address this sort of content.
One of the public figures who was the subject of the deepfake nude image was from India. The other was from the United States. Meta did not name the figures.
In the case of the Indian woman, a user reported the content to Meta, citing pornography. That report was closed as it was not reviewed within 48 hours. An appeal was made, but that was also automatically closed, meaning the image remained viewable. The user then appealed to the Oversight Board.
At that point, Meta reversed course and said the decision to leave the image up was a mistake and removed it.
The second image, of the American celebrity, showed her nude while a man groped her breasts. That image was removed promptly and added to the company’s automated enforcement system, which automatically removes images that break rules.
The board says it plans to assess whether current policies and enforcement practices are effective at addressing explicit AI images—and whether the company is enforcing those practices equally around the world.
Deepfakes are a fast-growing problem in general. A Channel 4 investigation in the U.K. found artificial images, many created by AI, of nearly 4,000 celebrities. And a 2019 report by the AI firm DeepTrace Labs concluded these images were overwhelmingly weaponized against women, with female actors and South Korean K-pop singers the most popular targets.
In January, a deepfake image of Taylor Swift went viral on Twitter/X, prompting the platform to block some searches for her name.
Meta’s Oversight Board is seeking public comments about a number of issues, including the harms posed by deepfake nude and pornographic images, how the company can address the issue, and the problems with automated systems that automatically close appeals before reviews can take place.
