Meta rolls out AI tools to catch ads that hide child abuse material

Meta announced Wednesday that more than 97% of the content it acted on in the first half of 2026 was flagged by its own systems before users reported it. In India, the company says it acted on 5.3 million pieces of child sexual exploitation content over the same period, with more than 98% caught before being reported.

Meta said the new tools target ads that look harmless but are designed to direct users to illegal content hosted outside Meta’s platforms. The company’s newest tool is a large language model system built to detect what it calls “signposting”: ads suspected of directing users to illegal content or other harmful activity elsewhere online — a tactic Meta says bad actors have increasingly adopted to evade detection. Because the ads themselves may not contain illegal material, Meta says it is now examining where an ad sends users, not just what the ad displays, so it can block the destination sites and act against the accounts running them.

Meta is also running additional AI-driven scans to catch child exploitation content that earlier systems missed, and says it will keep adding detection signals as it learns more about how these networks operate. A separate new tool, a “red-teaming AI agent,” probes Meta’s own safety measures for weaknesses that bad actors could exploit, which the company says could help it spot new abuse methods before they spread. Meta is also improving its ability to catch people who return to its platforms on new accounts after being removed.

The rollout comes as Meta faces continued pressure over the safety of children on its platforms, including lawsuits and criticism from lawmakers over the risks its social media services may pose to young users. In August, the company agreed to pay up to $18 billion to settle a child-safety lawsuit brought by 29 U.S. states. Meta has introduced several other child-safety features this year, including parental controls for Meta AI, preteen accounts on WhatsApp, and alerts to parents when children search Instagram for self-harm content. In September, WhatsApp added further parental controls letting parents limit teens’ use of Channels, control who sees their status updates, decide who can add their children to groups, and receive notifications about certain group activity.

Why it matters

The shift from screening ad content to tracking where ads send users could make it harder for bad actors to use seemingly harmless placements as a gateway to illegal material, as Meta continues to face legal and regulatory pressure over child safety on its platforms.