The UK government is expected on Monday to announce a ban on children under 16 using major social media platforms, in a policy described by Whitehall sources as “Australia-plus”. The platforms likely to be affected include TikTok, Snapchat, Instagram, YouTube, Reddit, Twitch, X, Threads, Facebook, and Kick — matching or exceeding the list banned in Australia when it introduced the world’s first outright under-16 social media ban in December 2025. The UK scheme would additionally restrict children from livestreaming on “safer” sites and from communicating with strangers on gaming apps, with social media curfews for under-16s and restrictions on AI chatbots reportedly under further consideration. A government consultation drew 116,000 responses, with 90% of parents backing the ban. Culture Secretary Lisa Nandy said that tech firms had “more than enough time to get their house in order.” Prime Minister Starmer framed the decision as a choice “about whose side we’re on.”
The received wisdom
The case for regulation is not absurd, and it deserves a serious statement before the critique. There is genuine and substantial evidence — from psychologists, from epidemiologists, from the lived experience of parents and children — that unrestricted adolescent exposure to algorithmically curated social media has contributed to rising rates of anxiety, depression, and self-harm in Western teenagers, particularly girls. The government’s consultation produced the figure of 83% of respondents saying that social media risks outweigh benefits for children. A democratic government that ignores a consultation majority of that size, particularly on child welfare, has a political problem as much as a policy problem. Australia’s ban, whatever its enforcement limitations, has forced a cultural conversation that framed the platforms as bearing civic responsibility for the content they serve to minors. That framing is not unreasonable. And the GCAP of child digital safety — where multiple countries are now converging on roughly the same policy toolkit — suggests this is a durable political consensus, not a passing moral panic.
A different read
The problem with this policy is not its aspiration but its mechanism and, more troublingly, the way in which the aspiration is used to foreclose harder questions about what actually makes children safer online.
Ian Russell, whose daughter Molly died aged 14 after encountering harmful content on Instagram, has called the ban “rushed” and warned it will “create a false sense of safety”. His is not a niche or ideologically convenient view: it is shared by the NSPCC, the Internet Watch Foundation, and Childnet — organisations whose entire institutional purpose is child protection. Their argument is not that social media is harmless; it is that an age-verification mandate drives younger users to less regulated corners of the internet, creates a black market in fake ID credentials, and provides parents and schools with a false reassurance that the problem has been solved. Culture Secretary Nandy acknowledged that “some young people will find workarounds” but argued the policy creates a worthwhile “cultural shift.” That is a very thin reed on which to rest a piece of legislation that will affect millions of children and require tech platforms to implement age-verification systems of unprecedented scope.
The age-verification problem is not merely technical; it is philosophical. Any system robust enough to reliably verify age would necessarily involve collecting more identity data from minors — and their parents — than most privacy advocates consider acceptable. The alternative, platform-side verification, has already been shown to be porous: Australia’s ban, in force since December 2025, has been accompanied by widespread reporting of easy workarounds. When platforms are asked to verify age without access to government identity databases, they are being asked to do something they cannot reliably do. When they are given access to such databases, the surveillance implications are substantial.
There is a broader pattern here worth naming. The British state has a recurring tendency, across Conservative and Labour governments alike, to reach for sweeping legislative mandates when confronted with complex social problems that defy neat solutions. The Online Safety Act, the Investigatory Powers Act, now this: each piece of legislation is presented as a decisive reckoning with a technology sector that has flouted its obligations. Each passes with broad parliamentary support and significant public approval. And each, in the years that follow, generates its own unintended consequences — overblocking, chilling effects on legitimate speech, enforcement gaps that the most vulnerable users fall through. The question is not whether the tech platforms have behaved irresponsibly toward children; they have. The question is whether age-based platform bans are the right instrument for addressing that irresponsibility, or whether they are a satisfying-feeling intervention that lets the underlying harms continue while giving politicians the ability to say they acted.
A more targeted and arguably more effective approach would focus on the specific harmful algorithmic features — infinite scroll, autoplay, recommendation systems optimised for engagement over wellbeing — rather than categorical access bans. These were among the options considered in the government’s own consultation and, reportedly, remain under consideration as supplementary measures. Making them the primary instrument, rather than the secondary one, would be harder to announce at a press conference. It would require sustained regulatory engagement with platforms rather than a clean legislative headline. But it would target the actual mechanism of harm rather than the age of the user.
What to watch
- Enforcement design: the specific age-verification methodology the government mandates, and whether it involves third-party identity data. Any proposal that requires platforms to collect government ID from under-16s will face legal challenges under GDPR and the UK GDPR.
- Australia comparison data: as the December 2025 Australian ban accumulates more longitudinal evidence, researchers will be able to assess whether youth mental health metrics or platform usage patterns changed measurably. This evidence will either vindicate or complicate the UK’s decision.
- Platform compliance and legal challenges: whether US-headquartered platforms challenge the legislation through trade agreement mechanisms (the UK-US trade relationship is already complex) or through domestic judicial review.
- The curfew and AI chatbot provisions: details expected “next month” on social media curfews and chatbot restrictions. These may prove more contentious — and more legally complex — than the platform ban itself.
— J