Meta announced Wednesday that it took action against 33.2 million pieces of child sexual exploitation content on Facebook and Instagram during the first half of 2026. Over 97 percent of this content was discovered by Meta systems prior to any user reports. In India specifically, Meta acted on 5.3 million pieces of similar content in the same period, with more than 98 percent caught before user reports.
Bad actors frequently use normal looking advertisements to route people toward illegal material hosted on external websites. This evasion tactic, known as signposting, means the initial ad contains no prohibited material itself. Meta deployed a new large language model system to analyze where advertisements send users rather than solely reviewing the ad content.
New systems target external links
This tracking allows Meta to block external websites that violate its rules and penalize the associated accounts. Meta also deployed additional artificial intelligence scans to catch exploitation content that older detection methods missed. The company stated it will continuously incorporate new signals as it learns how malicious networks adapt.
Another newly introduced system is a red teaming artificial intelligence agent that tests safety measures internally. This agent searches for vulnerabilities that bad actors could exploit to bypass platform protections. Furthermore, Meta is upgrading its detection systems to identify banned users who return to the platforms by creating new accounts.
Pressure mounts from lawmakers
These new defenses arrive as Meta deals with ongoing scrutiny regarding young user safety. The company faces active lawsuits and severe criticism from lawmakers concerning platform risks. Meta recently reached an agreement in August to pay up to $18 billion to resolve a child safety lawsuit involving 29 United States states.
Meta rolled out several safety features throughout the year. These updates include parental controls for Meta AI, restricted preteen accounts on WhatsApp, and parental alerts when children search Instagram for self harm content. WhatsApp also added parental controls in September to manage teenager channel usage, status visibility, group additions, and group activity notifications.
Meta rolls out more safety tools
Meta plans to continue adding new signals and security features as it adapts to evolving tactics used by malicious networks online.
