Meta’s Instagram has been serving up adverts in India that actively promote child sexual abuse material, a BBC investigation has found. The ads used explicit search terms including “rape” and “child video,” directing users straight to Telegram channels where the content was being distributed.
The BBC’s research team, working with child safety organisation Collective Shout, discovered that Instagram’s own ad algorithm was not only failing to block this content but was actively amplifying it to users who had engaged with related material. In other words, the platform’s recommendation system was doing the grooming for the abusers.
Meta, which owns Instagram, has long insisted it has zero tolerance for child exploitation on its platforms. It employs thousands of human moderators and uses automated detection tools. And yet, here we are.
When the BBC presented its findings to Meta, the company removed the adverts and said the content “violated its policies.” A spokesperson added that the company was “continuously working” to improve its systems. It’s the kind of corporate statement that lands with a thud when set against the reality of what investigators actually found.
“These weren’t hard-to-find dark web forums,” one researcher familiar with the investigation noted. “These were paid advertisements, on a mainstream platform, targeted algorithmically.”
India is Instagram’s largest market, with an estimated 362 million users in the country. The scale of that audience makes the failure all the more troubling. Telegram, meanwhile, has faced persistent criticism for becoming a haven for illegal content, with its light-touch moderation policies drawing fire from regulators across Europe and beyond.
The UK’s Online Safety Act, which received royal assent in 2023, places legal duties on platforms to proactively tackle child sexual abuse material. Meta is subject to Ofcom’s oversight here, and pressure is mounting on regulators to act decisively rather than wait for another BBC investigation to do the job for them.
The uncomfortable question this raises isn’t just about one set of adverts being taken down. It’s whether any platform operating at this scale can genuinely police itself, or whether governments need to stop trusting tech companies to mark their own homework.
