Meta uses new AI tools to detect ads tied to child abuse material

Meta said it took action against 33.2 million pieces of child sexual exploitation content on Facebook and Instagram in the first half of 2026, while launching new AI tools to detect ads and accounts that may secretly direct users to illegal child abuse material. More than 97% of the content was identified by Meta’s systems before users reported it.

A new large language model system is designed to identify “signposting,” a tactic in which ads appear normal but are suspected of sending users to illegal content or other harmful activity outside Meta’s platforms. Meta said it is now examining where ads send users, not only what the ads contain, so it can block rule-breaking destinations and act against the accounts behind them.

The company is also adding AI-driven scans to find content earlier systems may have missed and deploying a red-teaming AI agent to probe its safety measures for weaknesses. The rollout follows continued scrutiny of child safety on Meta’s platforms, including an agreement in August to pay up to $18 billion to settle a child safety lawsuit involving 29 U.S. states.