Meta Ran 332 Ads Using AI-Generated Child Sexual Abuse Imagery; 84 Targeted India: Report

Meta ran 332 paid ads carrying AI-generated child sexual abuse material on Facebook and Instagram over the past 10 months, including 84 in India, according to a report published on Tuesday by the Washington-based Tech Transparency Project, a research initiative of the non-profit Campaign for Accountability that seeks to hold large technology companies accountable.

The report comes two months after the Indian government ordered Meta to remove all ads and content promoting child sexual abuse material on Instagram after a BBC Eye investigation found such paid adverts were running in the country.

Most of the ads appeared recently. The Tech Transparency Project, or TTP, found that 274 of the 332 ads ran in August this year. It also found that 78 of the 84 India ads were published after the government order and a meeting with senior Meta officials, as reported by the BBC.

What Did The Report Find?

TTP said it identified 332 adverts featuring child sexual abuse material across Facebook and Instagram in the US, UK, EU, Australia and India. It said it had reported all the ads to NCMEC. Almost all of them, it said, began with a photo of a child that AI manipulated into graphic sexual acts.

The campaign group said it identified several photos of real children in the ads, including one of a member of a European royal family. It said those images were then manipulated using AI. It also said more than a dozen ads remained on Meta platforms even after it alerted the company to its findings.

One ad that TTP described as running in India showed a photo of a preteen girl animated into a graphic sex act, with a voiceover stating, "There are no restrictions. All the characters are real people and there are many options. This is the AI men actually use." The BBC reported it had not seen that ad. According to TTP, the same ad was posted by more than 50 different handles in India over a fortnight in August.

What Has Meta Said About The Findings?

Indian agencies started examining the issue after the BBC Eye investigation. They include the National Commission for Protection of Child Rights and the National Human Rights Commission.

The distribution of child sexual abuse material is a criminal offence in India. Meta's own policy also states that ads must not contain content that sexually exploits or endangers children.

Meta has said that when it becomes aware of apparent child exploitation, it reports it to the US-based National Center for Missing and Exploited Children, or NCMEC, to meet its legal obligations. The NCMEC is the centralised global reporting system for the online sexual exploitation of children. But Bhuwan Ribhu, founder of Just Rights for Children, a network of more than 250 organisations, said that does not satisfy Indian law.

Just Rights for Children has filed a petition in the Supreme Court citing the BBC findings. It has sought directions for mandatory reporting by social media platforms and for the identification of those promoting child sexual abuse material. "Social media companies are clearly flouting Indian laws by not reporting child sexual abuse content to Indian law enforcement agencies as mandated by our child protection laws," Ribhu said.

Meta defended its systems. In a statement to the BBC, the company said, "We don't tolerate nudify apps or any kind of child exploitation, whether real or AI-generated. These criminals constantly shift tactics to evade detection, which is why we keep strengthening our detection and enforcement."

In July, after the BBC investigation, Meta posted a detailed explanation on its website about how it fights child exploitation on its apps. "Anyone on our platforms can report ads if they believe they violate our policies," it said.

TTP said it was able to report 55 of the 84 India ads through Meta's own system because Meta only allows ads to be reported while they are live. For 43 of those, Meta replied: "We've taken a look and found that this ad doesn't go against our Advertising Standards." For 11, Meta informed TTP that the content had already been removed. One report had not received a response.