AI is turning ordinary photos of kids into sexual abuse imagery — here's how to protect your family

Parents often share photos of their children on social media, such as beach outings, playing soccer, or family gatherings. However, artificial intelligence (AI) has introduced a serious new risk - the potential manipulation of ordinary photos of real children into sexual abuse imagery without their knowledge or consent. This is not just a theoretical threat; research has uncovered hundreds of such ads on Facebook and Instagram.

The Tech Transparency Project (TTP) identified 332 AI-generated child sexual abuse material (CSAM) ads, depicting children whose regular photos had been altered by AI to appear as if they were engaging in sexual acts. The disturbing reality is that even an ordinary picture can serve as the raw material for creating convincing fake imagery with generative AI.

Researchers traced four minors back to their original sources, including teenage influencers with public social media accounts, a preteen available through a stock-photo website, and a young member of a European royal family. None of these children's original photos contained any sexual material; they were misused by converting them into sexual content through AI apps.

Once a photo is publicly accessible, anyone can download it and feed it into an AI tool designed to alter photographs. Meta's AI did not catch hundreds of these ads. The scale of the issue is significant, as many of these ads appeared after Meta introduced new AI technology to improve its detection capabilities. Meta has stated that it does not tolerate child exploitation, whether real or AI-generated, but bad actors deliberately attempt to evade their detection systems.

The ads were removed, but some reached a significant number of users before being taken down. The investigation reveals that AI can be used to create abusive material, with apps available through app stores and advertising platforms distributing ads promoting these tools. The problem extends beyond any single platform, as AI can create the imagery in the first place.

If you discover sexualized AI imagery involving your child, do not repost or widely share the image. Instead, preserve information about where the material appeared and report the content directly to the platform.