Meta Preps for US Midterms with Ad Ban, Scam Focus
Meta is activating its Election Operations Center for the 2026 U.S. midterm elections, with a focus on preventing scams, including those using AI-generated images of politicians. The company announced it will block new political advertisements during the final week of the campaign. The strategy reflects a shift toward pre-emptive moderation and managing the risks of AI in political discourse.
- Meta has invested over $20 billion in safety and security teams and technology since 2016, with around 40,000 people globally working on these issues. - The company's policy requires advertisers to disclose when AI is used to create or alter political or social issue ads that depict realistic people or events. However, advertisers are prohibited from using Meta's own generative AI tools for creating political ads. - While its overall safety and security staff is large, Meta's dedicated election integrity team was reportedly reduced from over 300 people during the 2020 U.S. election to approximately 60 people by mid-2022. - The one-week restriction on new political ads is a continuation of a policy used in the 2020 and 2022 U.S. elections. All authorized political ads are stored in a public Ad Library for seven years. - Since 2017, the company has removed more than 200 networks for "Coordinated Inauthentic Behavior" as part of its efforts to combat foreign and domestic influence operations. - In contrast to the temporary U.S. ad ban, Meta stopped accepting all paid political and social issue ads in the European Union in October 2025 due to the complexities of the EU's new Regulation on the Transparency and Targeting of Political Advertising (TTPA). - The Election Operations Center model has been previously deployed for other major elections, including those in Brazil, India, and for the EU Parliament, to coordinate experts in intelligence, data science, and security.
