Over the past nine months, Meta has run more than 50 paid advertisements containing explicit AI-generated child sexual abuse material and images of minors paired with sexually suggestive text, according to findings shared with WIRED by the Tech Transparency Project. The ads, discovered in Meta's public ad library, targeted users in the United States, United Kingdom, and over a dozen European nations. The listings show that some advertisements directed users to nudify or undressing applications.
The advertisements ran between November 2025 and early August 2026 across Facebook, Instagram, Messenger, and Threads, with Meta's ad transparency library showing that some ran for multiple days and certain ads specifically targeted male users. One video advertisement featured a thumbnail showing a child on the floor with overlay text reading "Realizing Deep Fantasies with Generation AI. There is so much more than what is shown, use your imagination." When clicked, the ad played clips of adults in sexual acts, then superimposed the child's face from the thumbnail onto a sex act. Another repeatedly posted ad displayed an image of a young girl in a sexual position with text stating "I can show you more." While many ads reached only a handful of accounts according to Meta's data, at least one reached 2,563 accounts across Europe, including France, Germany, Ireland, Italy, the Netherlands, Spain, Sweden, and the United Kingdom. Most ads were not actively showing to users when discovered but remained available in Meta's ad library for months.
"These ads made no effort to mask the images or hide what they were promoting," Katie Paul, director of the Tech Transparency Project, told WIRED. She emphasized that these weren't third-party posts but ads that Meta reviewed, approved, and allowed to run while collecting advertising revenue. The Tech Transparency Project reported its findings immediately to the National Center for Missing and Exploited Children's CyberTipline. Meta removed the ads after WIRED's inquiry, with a company spokesperson stating that many ads had minimal reach and were disabled before being shared, and that many predated new AI technology launched recently to detect and block violating ads at upload.
The findings reveal a significant gap in Meta's advertising approval systems, particularly concerning AI-generated abuse material. Meta's spokesperson noted the company removed over 36 million pieces of child sexual exploitation content last year and has taken legal action against nudify app developers. However, the discovery marks the second time in recent weeks that paid ads linked to child sexual abuse material have been found on Meta's platforms, suggesting ongoing challenges in screening advertised content before it reaches users. The fact that these ads remained in the company's public transparency library for months after running indicates detection failures extended beyond the initial approval process. Meta's statement pointed to recent improvements in AI-based screening technology, acknowledging that the systems require constant enhancement to address evolving threats from advertisers using generative AI to create abusive content.
The company says it will continue fighting this abuse through improved automated detection systems and enforcement actions. Meta's reliance on AI technology to identify violating content at the upload stage reflects an industry-wide shift toward preemptive screening as generative AI makes it easier to produce abusive material at scale. The challenge for platforms is clear: traditional content moderation approaches struggle when advertisers can rapidly generate variations of prohibited imagery that slip past automated filters designed to catch known abuse material. For companies that profit from advertising revenue while promising brand safety and user protection, failures of this magnitude carry both reputational and regulatory risks that extend well beyond individual enforcement actions.