Meta's Facebook and Instagram platforms carried more than 300 paid advertisements featuring child sexual abuse material (CSAM) this year, including AI-manipulated images involving real children, according to an investigation by the Tech Transparency Project (TTP).
The investigation found that several advertisements used photographs of real children sourced from social media posts and websites. The images were allegedly manipulated using artificial intelligence to depict children in sexually abusive situations.
In one case highlighted by TTP, a publicly released photograph of a minor member of a European royal family was allegedly manipulated into abusive material and appeared in an advertisement on Facebook and Instagram. The advertisement promoted a Chinese AI application and was reportedly targeted at adult male users.
According to TTP, the more than 300 advertisements identified in its investigation involved children ranging from toddlers and preteens to young teenagers. Many of the ads reportedly followed a similar format, combining adult-oriented content with AI-manipulated images of children.
TTP said it was able to identify multiple photographs of real children that had allegedly been used in the advertisements. The findings have raised concerns about the misuse of publicly available photographs and the ability of generative AI tools to transform them into abusive material.
The investigation also found that several advertisements were placed by Meta advertising partners in China, including a Chinese state-controlled company. Most of the ads reportedly directed users to AI applications developed in China, including apps offering or enabling so-called "nudification" features.
Meta did not initially respond to TTP's request for comment. However, after TTP shared its findings with the company, Meta carried out what the organisation described as a rapid cleanup of its advertising ecosystem.
According to TTP, around 150 advertisements that were still available in Meta's Ad Library were subsequently removed. Meta also updated records for more than 100 advertisements that had already been taken down, newly classifying them as violations of its child-exploitation policies.
TTP said the advertisements had previously been categorised under Meta's adult-oriented policies without explicitly identifying the involvement of children. The discrepancy has raised questions about whether existing moderation systems could be undercounting cases involving child exploitation, particularly when AI-generated or manipulated imagery is involved.
A Meta spokesperson said the company does not tolerate "nudify apps or any kind of child exploitation, whether real or AI-generated." The spokesperson added that many of the advertisements identified in the investigation had already been removed and that Meta continues to strengthen its detection and enforcement systems.
