Meta Deploys AI Tools to Catch Ads That Funnel Users to Child Abuse Material

Meta announced in October 2026 that it is rolling out new artificial intelligence tools designed to detect advertisements on Facebook and Instagram that appear harmless but are used to direct users to child sexual abuse material hosted elsewhere online.

The company said it took action against 33.2 million pieces of child sexual exploitation content on its platforms in the first half of 2026, with more than 97% identified by its own systems before any user reported it. In India alone, Meta acted on 5.3 million pieces of such content during the same period, with more than 98% detected proactively.

A key part of the new rollout is a large language model system built to identify what Meta calls “signposting” — ads that look ordinary but are suspected of directing users to illegal content or harmful activity on external sites. Meta said this tactic has emerged as bad actors adapt their methods to avoid detection. Because the ads themselves may not contain illegal material, Meta said it is now examining where an ad sends users, not just what the ad shows, allowing it to block offending destinations and act against the accounts behind them.

Meta is also deploying a “red-teaming AI agent” that tests its own safety systems for weaknesses that could be exploited. Additionally, the company is improving its ability to identify users who return to its platforms under new accounts after previous accounts have been removed.

The measures come as Meta faces ongoing legal and regulatory pressure over child safety. In August 2026, the company reached an agreement to pay up to $18 billion to settle a child safety lawsuit involving 29 U.S. states. Meta has also introduced parental controls for Meta AI, preteen accounts on WhatsApp, and Instagram alerts for parents when children search for self-harm content. In September 2026, WhatsApp added further parental controls over teenagers’ use of Channels, status updates, and group activity.