AI-Generated Child Sexual Abuse Ads Found on Meta Platforms

Researchers from the Tech Transparency Project (TTP) identified more than 50 image and video advertisements that contained AI-generated child sexual abuse material (CSAM) or minors shown alongside sexually suggestive text. The ads ran on Facebook, Instagram, Messenger and Threads and were targeted at users in the United States, the United Kingdom and over a dozen European countries. Over the past nine months the campaign reached several thousand accounts, and some ads linked to “nudify” or undressing applications. The listings were uncovered in Meta’s public ad-library tool, which catalogs paid content displayed on the company’s services.

Wired Magazine

The discovery marks the second instance in recent weeks that paid advertisements linked to CSAM have been located on Meta’s platforms. Earlier investigations had also revealed similar material, prompting renewed scrutiny of the company’s ad-review processes. The ads in question were not user-generated posts; they were approved and paid for through Meta’s advertising system, meaning they passed the company’s internal checks before being shown to audiences.

Wired Magazine

After the TTP shared its findings, Meta removed the identified ads, stating they violated the company’s policies on child sexual abuse and exploitation, as well as adult sexual solicitation and nudity. The removal was applied across all four platforms where the ads had appeared. Meta’s response emphasizes adherence to existing policy frameworks, though the incident highlights a gap in the pre-approval mechanisms that allowed the content to be published initially.

Wired Magazine

She characterized the episode as a failure of the company’s ad-review system, noting that the ads were allowed to run without interference while generating revenue. The watchdog’s assessment calls for stronger safeguards to prevent AI-generated CSAM from entering paid-media channels in the future.

Wired Magazine