केस विवरण
The Board has selected a case involving the posting by an activist of an Ethiopian intelligence service ID card, to explore Meta’s policies covering the sharing of potentially fake government IDs, which contain personally identifiable information in the context of raising awareness of crime and corruption. This case is an opportunity to address situations where people are concerned by the actions of law enforcement or those purporting to be law enforcement and what information they should be able to share to keep their communities safe.
In 2026, Ethiopian political activist Jawar Mohammed shared a post on his Facebook page in the Afaan Oromo language. In its caption Mohammed states that he has heard that an organized crime group called “120” has been terrorizing people in the Oromia region of Ethiopia, using National Intelligence and Security Service (NISS) ID cards to extort people. He names the leader of the gang and two individuals who he says participate in the extortions, one of whom is he says is a former police officer that now works for the gang. As “evidence”, the post includes both a picture of an Ethiopian NISS photo ID card, containing a name, role, photograph and phone number, and photos of two men. One of the men in the photos appears to be the same man shown on the ID card, the other is in what appears to be a military uniform. The content was posted in the context of widespread allegations against the “120” group and media and reporting. Mohammed’s page has over 3.4 million followers and the post received about 400 comments.
Three days after it was posted a team Meta had set up as part of its preparations for the June 2026 Ethiopian elections identified the post and referred it for expert review. Meta removed the post the following day for violating the Privacy Violations policy which prohibits the sharing of personally identifiable information (PII) as well as specifically prohibiting content sharing government IDs of law enforcement, military or security personnel. However, no strike or feature limit was applied, because it was unclear if the photo of the ID was authentic.
Mohammed’s appeal against the removal was closed after not being prioritized for human review. Mohammed then appealed the decision to the Oversight Board, stating that he had heard about the gang impersonating officers of the Ethiopian NISS, using fake IDs to extort money from people under threats of imprisonment or worse. He explained that, as part of his investigation, he obtained a copy of one of the fake IDs and confirmed with the relevant government authority that the name on it was not that of the person pictured, and that the ID had not been officially issued by them. He further explained that he posted the picture of the ID card as a public warning to alert citizens and to protect people from fraud and extortion. He added that the post was not meant to violate anyone’s privacy, rather to expose individuals using fake ID cards to impersonate intelligence officers to extort. The Board notes that, when users appeal to Meta, they cannot explain their reasoning or give any relevant context for the company to consider.
As a result of the Board selecting this case, Meta assessed its decision and maintained that the content violated the Privacy policy. However, as the violation was limited to the sharing of the ID card, the company was able to restore the rest of the post without the image of the ID. The company further explained that the policy would not prevent the user from posting the ID card with the PII redacted, or from posting a portion of the ID card with the PII not visible.
The Board would appreciate public comments that address:
- How social media platforms should address the competing interests of people sharing information about corrupt law enforcement officers and those impersonating law enforcement on the one hand, and the protection of legitimate law enforcement personnel, on the other.
- Whether Meta’s policies strike the right balance between allowing users to raise awareness around criminal activities and protecting the identity and personal information of people allegedly involved in these activities, particularly in crisis and conflict settings.
- How effective social media platforms are in moderating content in Ethiopian or other underrepresented languages, in particular when sharing public interest information about crime, scams and fraud.
- The role and shortcomings of mistake-prevention systems like cross-check in providing more nuanced review of content from human rights defenders and journalists, in particular around elections or crises in global majority regions.
- The role of appeals and due process in ensuring effective review of content moderation decisions, including cross-industry good practices for giving users more voice in their appeals.
In its decisions, the Board can issue policy recommendations to Meta. While recommendations are not binding, Meta must respond to them within 60 days. As such, the Board welcomes public comments proposing recommendations that are relevant to this case.
