Meta announced Wednesday that it is rolling out new AI tools to find ads and accounts designed to secretly direct people to illegal child abuse material. The company acted against 33.2 million pieces of child sexual exploitation content on Facebook and Instagram in the first half of 2026, with more than 97% of this content detected by Meta's systems before users reported it.
Meta's new large language model system detects "signposting"—ads that appear normal but are suspected of directing users to illegal content elsewhere online. Rather than only scanning what ads contain, Meta now examines where ads send users and can block websites that break its rules. The company is also deploying a "red-teaming AI agent" that tests Meta's own safety measures to find weaknesses bad actors could exploit.
Which age groups are affected
- Primary and secondary students who use Facebook, Instagram or WhatsApp are exposed to these ads, though Meta's detection aims to prevent access to illegal content.
- WhatsApp parents can now limit teenagers' use of Channels, control status visibility, and receive notifications about group activity.
- Meta AI parental controls and Instagram alerts for self-harm searches protect younger users across Meta's platforms.
How to help your child use Meta's platforms safely
- Enable parental controls on Meta AI and WhatsApp, then set limits on your teenager's Channels use and group additions.
- Review Meta's Instagram alerts feature, which notifies parents when children search for self-harm content.
- Check your child's account privacy settings and discuss which ads they encounter and where links direct them.