Key Takeaways
- Meta removed 4 million accounts and 36 million posts linked to child sexual exploitation in 2025
- The action follows Indian government pressure over Instagram ads allegedly promoting CSAM
- Meta is expanding AI detection systems to cover 98% of languages spoken online
Meta removed 4 million Instagram and Facebook accounts and 36 million pieces of content linked to child sexual exploitation last year. The company disclosed these figures in a blog post responding to Indian government pressure over Instagram advertisements that allegedly promoted child sexual abuse material.
The statement marks Meta's first direct response to a controversy that led India's Ministry of Electronics and Information Technology to summon the company. Union IT Minister Ashwini Vaishnaw had ordered MeitY to demand an explanation and ensure the offending ads were removed immediately.
What did Meta find in its investigation?
Meta claims it had already identified and disabled several violating ads and the accounts behind them before they were flagged publicly. The subsequent investigation led to additional removals, account bans, and blocking of URLs linked to policy-violating content.
“We're aware of recent news reports about Instagram ads in India that violated our policies against child exploitation. We take these concerns seriously, we never want this content on our platforms, and we're committed to improving our efforts to combat it.”
— Meta, Official Blog Post
The company pushed back against suggestions that its advertising systems deliberately targeted such content at users with inappropriate interests. Meta called these allegations "categorically inaccurate" and said it uses technology to identify accounts showing potentially suspicious activity related to children, automatically removing over 4 million such accounts last year.
How is Meta expanding its detection capabilities?
Meta is experimenting with advanced AI systems that cover languages spoken by 98% of people online. This represents a significant expansion from previous coverage of around 80 languages. The company says it has strengthened protections for teen users and increased collaboration with law enforcement agencies.
The 98% language coverage matters because CSAM networks often operate in regional languages to evade detection systems optimized for English and other major languages. If Meta's AI can actually detect coded language and imagery across this range, it would close a significant gap in content moderation.
Why does India's response matter for Meta?
India is one of Meta's largest markets, with over 500 million users across its platforms. Regulatory pressure from the Indian government carries real business consequences. The country has increasingly pushed back against tech platforms on content moderation, data localization, and competition issues.
The CSAM advertising controversy is particularly damaging because it touches Meta's ad business directly. If advertisers lose confidence that their brands won't appear alongside exploitative content, or if they fear regulatory backlash by association, Meta's core revenue stream faces risk.
What are the limits of Meta's response?
Meta's blog post addresses the symptoms but sidesteps a harder question: how did these ads get through in the first place? The company says it detected many violating ads before public reports, but clearly some made it onto the platform and ran long enough to trigger government intervention.
The 4 million removed accounts and 36 million removed posts are large numbers, but without context, they're hard to evaluate. What percentage of total CSAM-related content do these removals represent? How long did violating content stay up before detection? Meta doesn't say.
Logicity's Take
Meta's response follows a familiar playbook: release big enforcement numbers, announce AI improvements, and promise better collaboration with governments. The real test is whether the 98% language coverage actually catches coded CSAM terminology in regional Indian languages, not just Hindi. Companies running ad campaigns in India should audit where their ads appear and consider platform safety when allocating budgets. Tools like Brand Safety verification through third-party ad verification services may become more valuable as this scrutiny increases.
Frequently Asked Questions
How many accounts did Meta remove for child exploitation?
Meta removed 4 million Instagram and Facebook accounts linked to child sexual exploitation in 2025, along with 36 million pieces of related content.
Why did the Indian government summon Meta?
India's Ministry of Electronics and IT summoned Meta after reports that Instagram was carrying advertisements promoting child sexual abuse material. IT Minister Ashwini Vaishnaw directed the ministry to seek an explanation.
What is Meta doing to improve child safety detection?
Meta is expanding AI detection systems to cover languages spoken by 98% of people online, up from previous coverage of around 80 languages. The company says it has also strengthened teen protections and increased law enforcement collaboration.
Did Meta knowingly target CSAM ads at users?
Meta denies this, calling the allegation "categorically inaccurate." The company says it uses technology to identify and remove accounts showing suspicious activity related to children.
Need Help Implementing This?
If your organization needs guidance on brand safety, ad verification, or content moderation policies for your platforms, reach out to the Logicity team for consultation.
Source: Tech-Economic Times / ET
Huma Shazia
Senior AI & Tech Writer
Produced with AI assistance and reviewed by the Logicity editorial team. Learn more in our Editorial Policy.
