Meta, the owner of Facebook and Instagram, announced on Wednesday the implementation of new artificial intelligence tools to counteract advertising that covertly directs users to materials about child sexual abuse. As part of this announcement, the company revealed that in the first half of 2026 in India, it took action against 5.3 million materials related to child sexual exploitation.
According to the company's blog post, over 98% of content on Facebook and Instagram in India was detected and removed before users reported it. Globally, between January and June, 33.2 million such materials were processed across both platforms, with over 97% identified proactively.
Meta emphasized that maintaining the safety of youth on its applications is one of the most critical and never-ending tasks. The company noted that perpetrators have changed tactics by resorting to advertising that may appear harmless but redirects people to illegal content hosted off-platform.
After identifying this scheme, the company expanded the investigation beyond the advertised ads, blocked associated accounts, and stopped external links. New measures include a large language model-based system to detect 'signposting'—advertising content that seems innocent but is suspected of directing users to illegal content.
Furthermore, Meta improved the assessment of where advertising leads by analyzing not only the display itself but also the final destination, and added automated AI checks to identify materials that older systems failed to detect. An AI agent is also used to test its own safeguards for vulnerabilities, and the detection system for banned users creating new accounts has been strengthened.
In addition, Meta continues to use the PhotoDNA image and video matching technology, which it has used in its applications since 2011, and relies on behavioral signals and rule-based systems. The company shares hashes of identified materials with other organizations through the Lantern Coalition of Technologies program and blocks links to external websites hosting similar content, removing corresponding posts, comments, or ads.
Meta reports confirmed materials to the U.S. National Center for Missing & Exploited Children. In September, the company announced the direct transfer of child safety cases to the National Cybercrime Reporting Portal managed by the Indian Cybercrime Coordination Centre.
Meta investigators, including former FBI employees, are focused on dismantling networks of predatory accounts rather than working individually with each account. The company states that it constantly cooperates with authorities to hold offenders accountable.
India has increased control over social media platforms regarding harmful and illegal online content, including materials on child sexual abuse and deepfakes. In July, the government sent a notification to Meta concerning child sexual exploitation materials in paid Instagram advertising. The Ministry of Electronics and Information Technology (MeitY) immediately demanded that Instagram disable all advertising and content facilitating such access, requested a detailed explanation, and summoned its executives to India.
Subsequent intensive discussions with Meta's senior management led the Ministry to unequivocally state that such content constitutes a violation and cannot be protected under safe harbor status. The National Commission for Protection of Child Rights also summoned the CEO and Meta's head in India regarding the investigation into advertising with such material; the child rights body took note of the BBC Eye report on alleged advertising and requested clarification from Meta.
The BBC Eye investigation, published in early July, alleged that Instagram displayed paid advertisements promoting such content, some of which directed users to Telegram channels containing child pornography. The investigation also found that Meta's recommendation systems amplified the spread of videos with similar material.
