BBC investigation: Instagram published ads leading to child abuse content in India

An investigation reported by BBC Turkish has revealed that paid advertisements leading to child abuse material were shown to users in India on Instagram.

12punto

It has emerged that Instagram published paid advertisements in India that directed users to material related to the sexual abuse of children. According to the BBC Eye investigation, some of the ads used language suggestive of abuse and directed users to channels on Telegram.

Following the publication of the investigation, the Indian government requested that Instagram's parent company, Meta, immediately disable the ads in question. The government also demanded an explanation within one week regarding how these ads were permitted on the platform.

The BBC Eye investigation revealed that paid advertisements leading to child abuse material were published on Instagram.

It was stated that the ads seen by the BBC World Service directed users to Telegram channels, where child sexual abuse material was being sold for very low prices. The BBC reported the ads it identified and the associated Telegram channels to the authorities in India.

The report emphasized that some of the ads passed through Instagram's ad review system and were published. The BBC stated that when it reported one of the ads to Instagram, the platform responded 24 hours later claiming the content did not violate its "community guidelines."

META ANNOUNCED IT REMOVED THE ADS

In a statement to the BBC, Meta reported that the company had disabled some of the ads, suspended the accounts that published them, and removed additional ads following the BBC's findings. The company stated that URLs belonging to other content violating its policies had also been blocked.

In its statement, Meta said, "Child abuse is a heinous crime and Meta is working aggressively to combat it on our apps." The company described the allegation that it knowingly and intentionally targeted child-related ads to users interested in such material as "categorically false."

Meta argued that no system is perfect and that proactive detection technologies continue to work even after ads are published. The company stated that when cases of child abuse are identified, reports are made to the US-based National Center for Missing and Exploited Children (NCMEC) in accordance with legal processes.

Meta states that ads are checked by automated systems and, when necessary, human review before they are published.

It is stated that ads on Instagram are first reviewed primarily by automated moderation technology, and suspicious cases are directed to human review. Meta announced in March that it would reduce its reliance on human moderators and increase the use of artificial intelligence.

Meta generates the vast majority of its revenue from advertising.

Advertising is the primary source of income for Meta. The company announced that nearly 98 percent of its $200 billion revenue in the fiscal year ending in 2025 came from advertising. According to analysts, more than 90 percent of Instagram's revenue also consists of advertising.

WARNINGS FROM EXPERTS ON RESPONSIBILITY AND OVERSIGHT

Evaluating the BBC's findings, Madan Lokur, a retired judge of the Supreme Court of India, said the issue is very serious. Lokur expressed concern that Instagram is "making money by participating in criminal activity."

Madan Lokur said he was concerned that Instagram is "making money by participating in criminal activity."

While reminding that Indian law protects social media companies to a certain extent from content uploaded by users, Lokur stated that platforms cannot completely escape responsibility.

Former Facebook Vice President Brian Boland also told the BBC that he was "horrified but not surprised" by the findings. Boland argued that the Instagram algorithm is inclined to show more extreme and engaging content to keep users on the platform.

Brian Boland said the Instagram algorithm can suggest more striking content to keep users on the platform.

Boland stated that the problem is not that algorithms are designed to directly produce crime, but that systems driven by revenue and click targets can have dangerous consequences if not guided responsibly enough.

Telegram, for its part, announced that it uses both automated and human moderation to remove child sexual abuse material. The company reported that it removed more than 274,000 groups and channels linked to such material in 2026. While one of the two channels reported by the BBC was quickly shut down, it was reported that the other continued to upload new content for more than two weeks.

Dubai-based Telegram is not a member of NCMEC. The company joined the Internet Watch Foundation, which works with platforms to detect and remove child abuse material, in late 2024.

In India, both the distribution of child sexual abuse material and adult pornography are considered criminal offenses. According to the BBC, India ranked second after the US in 2025 with 1.9 million reports received by the NCMEC CyberTipline.

Shikha Goel says they receive more reports from Meta platforms compared to others.

Shikha Goel, Director of the Cybersecurity Bureau in Telangana state, said they receive more reports from Meta-owned Instagram and Facebook. Goel emphasized that this may not mean these platforms are the biggest source of the problem, as well-functioning detection systems can also generate more reports.

The Mumbai-based Rati Foundation stated that the vast majority of reports it receives regarding child sexual abuse material come from Meta platforms. Siddharth Pillai, co-founder of the foundation, said that criminals can evade oversight efforts by using the transition from Instagram to Telegram and re-uploading removed content.

Bhuwan Ribhu, who works in the field of children's rights, also drew attention to the organized crime dimension of the problem. According to Ribhu, monitoring the demand and supply chain, cross-border intelligence sharing, and international cooperation are of critical importance in the fight against online child abuse.