Meta Approved Dozens of Ads Containing AI-Generated Child Sexual Abuse Imagery, Watchdog Finds

Meta ran paid ads containing AI-generated child sexual abuse material for months before an outside watchdog caught them, according to details shared with WIRED.

Researchers at the Tech Transparency Project (TTP), an independent watchdog group, found more than 50 image and video ads in Meta's own ad library that violated the company's policies against child sexual exploitation and adult sexual content. The ads ran across Facebook, Instagram, Messenger, and Threads between November 2025 and the start of August 2026, according to WIRED.

Some of the ads reached several thousand accounts before disappearing, per Meta's own tracking data. They targeted users in the United States, the United Kingdom, and more than a dozen European countries. Some ads specifically targeted men.

One video ad flagged by researchers used a thumbnail of a child sitting on the floor with text reading, "Realizing Deep Fantasies with Generation AI. There is so much more than what is shown, use your imagination." According to TTP, clicking the ad played footage of adults engaged in sexual acts with the child's face from the thumbnail superimposed onto a sex act. Another ad depicted a young girl in a sexualized pose with the caption "I can show you more."

"These ads made no effort to mask the images or hide what they were promoting," TTP director Katie Paul told WIRED. "It's important to point out that this isn't content posted by third parties on Facebook or Instagram, these are ads that were reviewed, approved, and allowed to run by Meta, never encountering interference while the company collected the ad dollars."

That distinction matters. Meta's ad system requires human and automated review before an ad goes live, unlike ordinary posts, which get far less scrutiny. These weren't ads that slipped past moderators after the fact. Meta's own process cleared them.

Meta pulled the ads after WIRED brought them to the company's attention and said it continues to fight this kind of abuse. "Sexual exploitation is horrific, and we work aggressively to keep it off our platform," the company said in a statement. "The majority of these ads had minimal reach and many were disabled before WIRED shared them. Many of these ads also predate new AI technology we launched recently to better detect and block violating ads at upload, and we're constantly improving these systems. From removing over 36 million pieces of child sexual exploitation content last year to taking legal action against nudify app developers, we will continue to relentlessly fight this abuse."

That defense deserves a fair hearing. Meta operates at a scale most companies never will, billions of ad impressions across four platforms, and it does report violations to the National Center for Missing and Exploited Children. Automated detection systems, even good ones, will miss things, especially as AI-generation tools evolve faster than moderation tools can adapt. Meta says newer detection technology launched after some of these ads were originally submitted. It's worth taking seriously.

But it doesn't erase the core problem: this is not the first time in recent months that paid ads tied to child sexual abuse material have surfaced on Meta's platforms, and the pattern points to a system that keeps failing at the one job that should never fail.

The BBC reported in July that Instagram ran ads for child sexual abuse material in India. And a separate investigation, reported by the Japan Times in late July, found that Meta's advertising partner in China, a Beijing-based firm called GatherOne, had delivered roughly 7,600 ads for AI "nudify" apps that let users transpose women's faces onto naked bodies, according to TTP. One app flagged in that report has been linked by users to material some described as child pornography, per the Japan Times report.

Regulators, victims, and families have pushed tech companies for years to crack down harder on deepfake porn services and nudify apps, according to the Japan Times. The Take It Down Act, a U.S. law referenced in that reporting, criminalizes publishing nonconsensual sexual content and requires platforms to remove it. But TTP's findings suggest enforcement on Meta's own ad platform, the ads it directly profits from, has lagged well behind the law and the company's own stated policies.

Meta's ad library is a public transparency tool, meant to let anyone see what's running and who it's targeting. That same tool is how outside researchers, not Meta's internal systems, caught these ads. Some had been sitting there, unflagged, for months.

No regulatory action or investigation tied specifically to this batch of ads has been announced. Whether the Federal Trade Commission, European regulators, or Congress respond to a now-repeated pattern of CSAM-linked advertising on Meta's platforms remains an open question. Meta has not said how many total ad dollars it collected from the specific ads TTP identified, or whether it will refund or audit the advertisers behind them.