Nine months after the Online Safety Act’s child protection duties came into force, the proportion of British children encountering suicide material, hardcore pornography, and violent content on social media has barely shifted. Ofcom’s own review, published Friday, confirms what campaigners have warned since royal assent: the architecture of harm remains intact, and the algorithms driving it are still running at full throttle.
The regulator’s assessment of the first nine months of the regime — covering July 2025 to March 2026 — found “no significant change in the proportion of children encountering harmful content.” Not a dip. Not a statistically meaningful reduction. The severity of the harm children report — fear, confusion, distress — is likewise unchanged.
The Damning Verdict
Dame Melanie Dawes, Ofcom’s chief executive, put it bluntly to MPs this week: “We do not see a safety culture from these large social media companies.” Her agency’s research tracked exposure among eight- to seventeen-year-olds across user-to-user services. It mapped the impact on wellbeing. It analysed the safety tools platforms claim to have built. The conclusion is stark. Personalised recommendation feeds — the “For You” algorithms that power TikTok, Instagram, and the rest — were “repeatedly found” to be the primary vector for the continuing deluge.
Girls bear the brunt. They are most likely to encounter suicide and eating disorder content, served up by the same recommendation engines that push pornographic material into their feeds. The result, Ofcom found, is confusion and upset at an “alarming rate,” to borrow the NSPCC’s phrase. Rani Govender, the charity’s head of child safety online, didn’t mince words: “Tech companies have had months to act, yet too many platforms are still built to keep children hooked with their algorithms continuing to push harmful content to their young users.”
Boys face a different pipeline. Ofcom documented how young men and boys consuming “motivational and self-improvement content can rapidly become exposed to more extreme content which includes the promotion of misogyny and racism.” Misspelled hashtags. Coded language. The moderation systems — where they exist — are being gamed in real time.
Algorithms: The Engine of Harm
The coroner’s conclusion in the Adriana Retyte inquest landed just a week before this report. The fourteen-year-old Langley School pupil took her own life in May. The coroner found that access to suicide-related material on TikTok “may have played a significant part” in her death. A day later, Clavicular — the twenty-year-old influencer who promotes surgical and pharmaceutical body modification as part of extreme male beauty regimes, and who has appeared alongside Andrew Tate — pleaded not guilty to rape in a US court. The “manosphere” content ecosystem, supercharged by recommendation algorithms, is not a theoretical risk. It is a lived reality for the children Ofcom is mandated to protect.
Ian Russell has lived this reality for eight years. His daughter Molly was fourteen when she died in 2017 after viewing suicide material on social media. He watched the Online Safety Act crawl through Parliament. He watched the duties commence. He watched the nine months tick by. “Staggering but sadly not surprising,” he told me of Ofcom’s findings. “It is time that Ofcom stood up for children and investigated the major flaws in TikTok and Instagram’s design features, not just small technicalities that have been their focus so far.”
His frustration is shared across the sector. Mitali Sud, senior policy manager at Internet Matters, warns that the proposed Australian-style ban on social media for under-sixteens — due to take effect next spring across Instagram, YouTube, TikTok, Snapchat, Facebook and X — cannot become a substitute for enforcement. “Children will continue to face risks on services covered by the act but not the proposed ban, including gaming and messaging platforms,” she said. “Sixteen- and seventeen-year-olds will still use social media. A social media ban cannot replace the need to make online services safer for children.”
Regulatory Paralysis and Legal Warfare
Ofcom insists progress is being made. Widespread age checks. Stronger anti-grooming protections. Tougher safeguards against child sexual exploitation and abuse. The spokesperson pointed to reductions in harmful content on Minecraft, Call of Duty, YouTube, Pinterest and Facebook. An investigation into TikTok over child safety concerns launched in July. But the regulator is fighting a war on two fronts. Meta, TikTok, and X — owned by Elon Musk’s X Corp — have launched legal challenges to Ofcom’s powers. Dawes has warned privately that the watchdog risks being “absolutely clobbered legally by deep-pocketed individuals and firms.”
The government, for its part, is doubling down on the ban. A spokesperson said platforms “have both a moral and legal duty to protect children and we expect them to work with us to set a new normal for future generations.” But Lisa Nandy, the Culture Secretary, now holds an assessment that tells her the current normal is failing. The data shared with her department does not even cover the period from this spring, when Ofcom says it “stepped up pressure” on Facebook, YouTube, Snap, Roblox and TikTok.
Why it Matters
A generation of children has been used as beta testers for an unregulated industry for over twenty years. The Online Safety Act was supposed to draw a line. Instead, nine months in, the line has not moved. The algorithms that serve suicide content to a grieving fourteen-year-old, that funnel misogyny to a lonely thirteen-year-old boy, that push hardcore pornography to an eight-year-old girl — they are still the core product. Not a bug. A feature. Until the design changes, the law is just text on a page, and the coroners will keep writing the same conclusions.
