Meta rolls out new AI tools to detect ads that secretly lead to child sexual abuse material
Meta launches new AI tools after discovering ads on its platforms that may look normal but direct users to harmful content elsewhere online.
TL;DR
- Meta removed 33.2 million pieces of child sexual exploitation content from Facebook and Instagram in the first half of 2026.
- New AI tools are being rolled out to detect ads that secretly direct users to illegal child abuse material.
- Over 97% of the detected child exploitation content was found by Meta's systems before user reports.
- A new LLM system called 'signposting' detects ads that appear normal but may lead to illegal content elsewhere.
- Meta's AI will now examine ad destinations, not just their content, to block rule-breaking sites and accounts.
- Additional AI scans are being used to find previously missed child exploitation content.
- A 'red-teaming AI agent' tests Meta's safety measures for weaknesses.
- Systems are being improved to find users who return with new accounts after previous ones were removed.
- Meta is facing pressure over child safety, with a recent $18 billion settlement in a child safety lawsuit.
- Recent child-safety features include parental controls for Meta AI and WhatsApp group restrictions for teens.