Meta introduces AI tools to block ads that secretly route users to child abuse content

Meta announced on October 7, 2026, a new set of AI tools aimed at identifying and blocking ads and accounts that could secretly steer users to child sexual abuse material hosted outside its platforms. They include a large language model-based system to detect "signposting" ads, and a "red-teaming AI agent" designed to test Meta's own safety measures and look for weaknesses that bad actors could exploit.

Meta said signposting ads may look normal but are suspected of directing users to illegal content or other harmful activity elsewhere online. While the ads themselves may not contain illegal material, they can act as a gateway to websites hosting child sexual abuse material. The company said it is now looking at where an ad sends users, not just what the ad contains, and can use that information to block violating destinations and take action against the accounts behind them.

The company is also running additional AI-driven scans to find child exploitation content that earlier systems may have missed, and is improving detection of users who return to its platforms with new accounts after their previous accounts were removed.

In a related announcement, Meta said it acted on 33.2 million pieces of child sexual exploitation content on Facebook and Instagram in the first half of 2026, with more than 97% detected by its systems before users reported them. In India, Meta acted on 5.3 million pieces of such content during the same period, with over 98% found proactively.

The new tools arrive as Meta faces continued scrutiny over child safety. In August 2026, the company agreed to pay up to $18 billion to settle a child safety lawsuit brought by 29 U.S. states. This year, Meta has introduced several child-safety features, including parental controls for Meta AI, WhatsApp accounts for preteens, Instagram alerts for parents when children search for self-harm content, and additional WhatsApp parental controls in September.