Non-Public Figures Need More Protection Over Sexualized Deepfakes
23 de junio de 2026
The Oversight Board has overturned Meta’s decision to leave up on Instagram a reportedly AI-generated video impersonating a woman and calls on Meta to strengthen protections for non-public figures who are the targets of sexualized deepfakes.
Why This Matters
The Board’s decision and recommendations come at a time when governments, such as in India, the UK and Spain, are setting up new rules for platforms on AI-generated content. The European Union (EU) has reached an agreement to prohibit AI systems that generate non-consensual sexually explicit and intimate content or child sexual abuse material, such as AI ‘nudification' apps. Other platforms, such as X, are facing scrutiny over sexually explicit images made by AI chatbots.
It is clear that the scale, speed and sophistication of AI tools have resulted in a proliferation of AI-generated sexualized non-consensual content globally. The spread of sexualized deepfake videos leads to reputational and psychological harm, which disproportionately impacts women and girls, and has a chilling effect on participation in social and political life.
About the Case
In September 2025, an Instagram user posted a reportedly AI-generated eight-second video of a woman adjusting her form-fitting dress and moving her body, with her underwear visible in a few frames.
The next day, Meta’s automated system that detects content that may pose harm to individuals and has a high likelihood of virality identified the post, but it was not prioritized for human review. A few days later, two users reported the content, but it remained on the platform. One of the users appealed to Meta, but the post was not reviewed, and the user then appealed to the Board. The user who appealed told the Board that the video was AI-generated and that it impersonated a friend (who had already closed her account) of theirs without the friend’s consent.
When the Board brought the case to Meta’s attention, the company’s subject matter experts reviewed the post and concluded that it did not merit removal under the company’s Community Standards but made the post visible only to adults.
Key Findings
The Board finds that the post violates the prohibition against sharing non-consensual intimate imagery (NCII) under Meta’s Adult Sexual Exploitation Policy. This is because the post fulfills all three of the policy’s criteria for non-consensual intimate imagery – it appears to be non-commercial content in an apparently private setting; the woman depicted is “near nude”; and there is a lack of consent.
Meta said that when the content was flagged, the company had no indication that the individual depicted in the video was “a real person” because they did not report the content. Among the signals Meta uses for determining lack of consent is a report from the person depicted, in addition to captions, comments or titles that suggest a vengeful context; or reports from independent sources such as law enforcement, media or representatives of survivors of non-consensual intimate imagery.
The Board finds that AI-generated impersonation is non-consensual by default and should be added to the set of signals the company uses to establish lack of consent.
Broadening the signals of lack of consent in this way would especially benefit non-public figures who are the targets of non-consensual intimate imagery because it would reduce the burden on victims to report the abuse themselves. Experts the Board consulted during the case said victims benefit most from rapid removal and better ways to flag content that do not rely on self-reporting.
To further reduce this burden, the Board finds that Meta should allow verified accounts from friends and families of those targeted to report violating content on their behalf.
Meta should also improve the process for reporting and appealing non-consensual intimate imagery globally on its platforms. Meta and other large language model developers should embed safeguards in system design and implement content credentials at scale.
The Oversight Board’s Decision
The Board overturns Meta’s decision to leave up the content and requires that the post be removed.
The Board recommends that Meta:
- Should add a new signal for lack of consent in the Adult Sexual Exploitation policy: context that content is AI-generated sexualized impersonation of real people.
- Allow users to designate “connected accounts,” such as trusted friends, family members or associates, that can report on their behalf potential Community Standards violations involving non-consensual intimate imagery, including impersonations.
- Include AI-generated sexualized impersonation as a separate category in standard content reporting and appeal forms, distinct from “harassment” or “nudity.” The reporting forms should be made available globally.
The Board also reiterates the following recommendations from earlier decisions:
- Change the word “derogatory” in the prohibition on “derogatory sexualized photoshop” to “non-consensual” and replace the word “photoshop” with a more generalized term for manipulated media.
- Implement content credentials (as laid out by the Coalition for Content Provenance and Authenticity (C2PA)) at scale and ensure that they are clearly and consistently visible and accessible to users whenever the provenance details are available.
Futher Information
To read public comments for this case, click here.
