Reported AI-Generated Sexualized Video
UPDATED SEP 22, 2026
Standard Case 2026-021-IG-UA
Today, February 12, 2026, the Oversight Board selected a case appealed by an Instagram user regarding a post that contains a sexualized, AI-generated video impersonating a woman. The video includes a woman in a form-fitting dress adjusting her dress and moving her body, with her underwear visible in a few frames.
Meta determined that this content did not violate our policies on Adult Nudity and Sexual Activity, as laid out in the Community Standards, and left the content up. In line with our policies, Meta determined that the content should only be visible to adults. We restrict the visibility of “photorealistic or digital imagery of persons where crotch, buttock or female breast(s) are the focus of the image” to users aged 18 and up.
We will implement the Board's decision once it has finished deliberating, and will update this post accordingly. Please see the Board's website for the decision when they issue it.
Case decision
We welcome the Oversight Board's decision today, June 23, 2026, on this case. The Board overturned Meta’s original decision to leave up the content.
Meta will comply with the Board's decision within 7 days.
When it is technically and operationally possible to do so, we will also take action on content that is identical and in the same context. For more information, please see our Newsroom post about how we implement the Board’s decisions.
After conducting a review of the recommendations provided by the Board, we will update this post with initial responses to those recommendations.
Recommendations
Recommendation #1 (Assessing feasibility)
To help ensure violating content is removed, 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.
Our commitment: In prior responses, we have shared details about the nuances between how our Adult Sexual Exploitation (ASE), Adult Nudity & Sexual Activity (ANSA), and Bullying & Harassment (B&H) policies address potentially sexualized content. We are continuing to explore ways to share that guidance with our internal teams and make updates to ensure we are addressing this content when there are indicators that it is AI-generated. Functionally, however, we do not believe that the signal proposed by the Board is operationally feasible for the structure and intended purpose of the non-consensual intimate imagery (NCII) policy. We will continue to explore policy and other approaches to address this content in line with the Board’s related recommendations from prior cases and the recommendations below.
Considerations: As the Board highlights in their decision, under our ASE policy, we remove non-consensual intimate imagery (NCII) when it appears to be non-commercial content in a private setting, the woman depicted is “near nude”, and there is a lack of consent as established through a list of criteria. We recently also updated our ASE policy to explicitly explain our approach to removing “nudify” applications. Separately, under our B&H policy, we remove derogatory sexualized photoshop, or drawings, which includes imagery created with generative-AI. Finally, under our ANSA policy, we have a number of other levers, including removal and age-gating certain types of content, that may be sexual or involve nudity.
That being said, sharing AI-generated imagery that may include near nudity and may be non-commercial or in a private setting does not always equate to a lack of consent or intent to share NCII. Moreover, adding a signal that includes treating AI–generated content as a lack of consent indicator in certain cases would not be a proportionate response to posting this content. We strictly enforce on NCII content due to the high levels of harm involved. Incorporating this additional signal may result in overenforcing on content that is being shared consensually. For example, there may be instances whereAI imagery is shared with near nudity or is shared in an affirming manner by the person depicted, and enforcement under NCII would not be a proportionate response in instances where content should instead be age-gated so that users under 18 do not see the content rather than outright removing it.
As a result, we will continue to remove NCII if it meets our existing signals for this type of content and will continue to assess this content under our other policy lines, including derogatory sexualized photoshop. As we work through additional guidance around other related policies and new approaches to enforcing on this content as recommended below, we will consider if there are other ways that meet the spirit of this recommendation. We will provide updates on this work in future reports to the Board.
Recommendation #2 (Assessing feasibility)
To reduce the burden on victims of non-consensual intimate image abuse, Meta should allow users to designate connected accounts, such as those of trusted friends, family members or associates, that can report on their behalf potential Community Standards violations involving non-consensual intimate imagery, including impersonations.
Our commitment: We acknowledge the importance of reducing the burden on victims of non-consensual intimate image (NCII) abuse and have a number of tools to address NCII. We are assessing the feasibility of allowing users to designate connected accounts to report on their behalf, while continuing to expand existing tools and partnerships that reduce reliance on direct victim reporting.
Considerations: We recognize the importance of reducing the burden on victims of non-consensual intimate image abuse and are assessing the feasibility of expanding pathways for trusted parties to report on their behalf.
We have invested significantly in tools and partnerships that help victims seek protection without needing to report each instance of content directly on our platforms. We supported the development of StopNCII.org, which helps prevent the spread of adults' real and AI-generated intimate imagery by allowing individuals to create hashes-essentially digital fingerprints- that participating platforms like Meta use to proactively detect matching content. We also supported the National Center for Missing & Exploited Children (NCMEC) in developing Take It Down, which provides similar protections for young people. Additionally, we accept reports from law enforcement, media organizations, and trusted partners as credible indicators of non-consent under our Adult Sexual Exploitation policy.
Regarding the specific proposal to allow users to designate connected accounts to report on their behalf, we are continuing to assess this approach, consistent with our response to recommendation #3 in the Gender Identity Debate Videos case. A primary challenge remains creating a system that can verify, at scale, that a reporting person is genuinely authorized to act on behalf of the depicted individual. Without reliable verification, such a system risks misuse through false claims against consensually shared content. We are exploring how existing tools and partnerships could be expanded to address this, and will provide updates in future reports to the Board.
Recommendation #3 (Assessing feasibility)
To ensure effective and accessible reporting pathways to report on AI-generated non-consensual intimate imagery abuse, Meta should include AI-generated sexualized impersonation as a separate category in standard content reporting and appeal forms, distinct from "harassment" or "nudity." These uniform reporting forms should be available globally.
Our commitment: We will assess the feasibility of including AI-generated sexualized impersonation as a distinct category within our reporting and appeal flows. This assessment will be informed by ongoing work to improve reporting accuracy and by related policy updates responding to the Board's recommendations in the Explicit AI Images case.
Considerations: We are committed to ensuring that reporting pathways are effective and accessible for users experiencing harm from AI-generated non-consensual intimate imagery. Today, users can report this content through several pathways depending on the nature of the violation, including under our Adult Sexual Exploitation policy (which covers NCII, including AI-generated content), our Bullying and Harassment policy (which addresses derogatory sexualized manipulated imagery), and our Adult Nudity and Sexual Activity policy. Additionally, we have developed dedicated reporting pathways for intimate image abuse, as well as partnerships with StopNCII.org and Take It Down that allow users to seek protection proactively.
We recognize the Board's concern that users may face difficulties identifying the correct reporting category when the harm they experience is specifically AI-generated sexualized impersonation. We are actively working to improve how users classify harms within our reporting flows more broadly, and will evaluate whether a distinct category for AI-generated sexualized impersonation could improve both reporting accuracy and the user experience. This assessment will need to account for how such a category would align with our existing policy architecture, reviewer training, and related updates currently in progress in response to the Board's Explicit AI Images recommendations. We will provide updates on this work in future reports to the Board.