Meta Unveils New AI Tools to Tackle Child Sexual Exploitation Online

Meta has introduced a new set of artificial intelligence-powered measures to strengthen its efforts against child sexual exploitation on Facebook and Instagram, reporting that it took action against 5.3 million pieces of such content in India during the first half of 2026.

The company said more than 98 per cent of the violating content removed in India between January and June was detected proactively by its systems before users reported it.

“Between January to June 2026, we actioned 5.3 million pieces of child sexual exploitation content on Facebook and Instagram in India, with over 98 per cent found and proactively addressed before anyone reported it,” Meta said.

Globally, the company said it acted against 33.2 million pieces of child sexual exploitation content across Facebook and Instagram during the same period. More than 97 per cent of that content was identified proactively before being reported by users.

Meta said online predators are increasingly adopting sophisticated methods to avoid detection. These include advertisements designed to look harmless while covertly directing users to illegal material hosted outside Meta's platforms.

To counter such tactics, the company has strengthened its advertising review systems and added new AI-based detection capabilities.

Among the new measures are large language model-based systems designed to detect “signposting” content. Such material may appear innocuous on the surface but can potentially direct users towards child sexual exploitation material or other harmful activities.

Meta has also enhanced its ability to examine the destinations associated with advertisements. The company said it can identify and block links that lead to violating content and take action against the accounts associated with them.

The company said that after blocking a link, its systems search for and remove other material containing the same link, including advertisements, posts and comments. Meta also takes steps to prevent users from subsequently posting content containing blocked links on Facebook, Instagram and Threads, while advertisements containing such links are rejected during the upload process.

The latest measures are part of Meta's broader effort to use AI and automated detection systems to identify harmful content and prevent it from spreading across its platforms.