- Meta announced advanced AI systems and updated ad reviews to combat child sexual abuse content
- India saw 5.3 million CSAM pieces actioned on Facebook and Instagram from Jan to June 2026
- Over 98% of violations in India were detected proactively by automated Meta systems
Social media giant Meta announced on Wednesday a suite of enhanced measures, including advanced artificial intelligence (AI) detection systems and revamped ad review protocols, designed to combat child sexual abuse material (CSAM) across its ecosystem. The Silicon Valley conglomerate revealed that in India alone, it actioned 5.3 million pieces of child sexual exploitation content across Facebook and Instagram between January and June 2026. Over 98% of these violations were detected and addressed proactively by automated systems before any user reported them.
Alongside the enforcement figures, Meta confirmed it is establishing direct reporting integrations with the Indian Cybercrime Coordination Centre (I4C) under the Ministry of Home Affairs. Suspected child safety and CSAM cases originating in India will now be routed directly through the National Cyber Crime Reporting Portal (cybercrime.gov.in), bypassing traditional diplomatic or international intermediary bottlenecks to give domestic law enforcement faster access to actionable intelligence.
READ | Meta Pushes Back On Muse Privacy Claims, Says Message Access Requires Permissions
Also Read | Layoffs Are Rising In Tech, But These Skills Are Helping People Unlock Better Pay
AI-Driven Ad Review Overhaul
"Today we're sharing the additional measures we've put in place to strengthen our efforts to combat child sexual exploitation on our platforms," it said. These measures include new AI tools to address the identified adversarial tactics and improve the efficacy of systems for detecting ads that either feature child sexual abuse material or that direct people to violating activity off-platform. "Keeping young people safe on our apps is among the most important work we do, and it never stops," Meta said.
It further noted that predators have shifted tactics to advertising, where an ad can look harmless on its own while leading people to illegal material hosted off its platforms. The policy and product updates focus heavily on tightening Meta's commercial ad pipeline, targeting bad actors who use sponsored posts to promote illicit content or funnel users to violating external sites. The key technical enhancements detailed in Meta's safety roadmap include:
- Large Language Model (LLM) Detection: Deploying advanced text models to parse nuanced keywords, coded language, and evasive emoji patterns across ad copy.
- Ad Destination Parsing: Scanning landing pages, redirected URLs, and external web environments to catch ads that direct users to violating third-party activity off-platform.
- Proactive AI Sweeps: Running automated, platform-wide sweeps to surface hidden or newly circulating abusive networks.
- Red-Teaming with AI Agents: Utilizing adversarial AI agents that deliberately challenge internal defenses, surfacing system blind spots before predators can exploit them.
- Recidivism Detection: Upgrading account-linking mechanisms to block previously banned bad actors from creating fresh ad accounts or rotating digital payment profiles.
READ | Mark Zuckerberg's $4 Billion Lab Experiment: Meta's AI Tax Dodge Explained
Direct Enforcement Channel with India's I4C
India represents one of Meta's largest user bases globally, making domestic compliance and rapid escalation critical. Historically, global platforms reported such violations predominantly to the U.S.-based National Center for Missing & Exploited Children (NCMEC), which then relayed leads internationally.
Besides, Meta said it continues to use PhotoDNA and other image and video matching technology, which it has deployed across its apps since 2011, along with behavioural signals and rules-based systems.The transition to direct reporting via I4C's National Cyber Crime Reporting Portal creates an immediate bridge to Indian central and state cyber police cells, speeding up the rescue of victims and the apprehension of perpetrators.
Meta reiterated that its advertising policies enforce zero tolerance for child endangerment, noting that the combination of continuous model retraining and direct law enforcement pipelines aims to close gaps exploited by organized predator networks. India has stepped up scrutiny of social media platforms over harmful and illegal online content, including child sexual abuse content and deepfakes. In July, the government issued a notice to Meta on child sexual exploitation and abuse material in paid advertisements on Instagram.
The Ministry of Electronics and Information Technology (MeitY) immediately ordered Instagram to disable all ads and content promoting and facilitating access to such material, demanded a detailed explanation from the company, and summoned its global executives to India. Multiple and intense rounds of discussions with Meta's top executives followed, in which the Ministry made it unequivocally clear that such content constitutes a violation and cannot be covered by safe-harbour protection.
Essential Business Intelligence, Sharp Market Insights, Practical Personal Finance Advice, Daily Fuel, Gold and Silver Prices and Latest Stories — On NDTV Profit.
