On June 15, 2026, Prime Minister Keir Starmer announced that Britain would bar under-16s from major social media platforms — TikTok, Snapchat, Instagram, Facebook, YouTube and X — while exempting messaging services like WhatsApp and Signal. "This is a line in the sand," Starmer told reporters. "Tech giants had their chance and failed." The mechanism is section 70 of the Children's Wellbeing and Schools Act 2026, which inserts a new section 214A into the Online Safety Act 2023, letting ministers impose the restriction through secondary legislation rather than a fresh bill. Draft regulations go to Parliament before Christmas 2026; enforcement begins spring 2027. Alongside the ban, the government will require livestreaming and stranger-contact features to be switched off by default for 16- and 17-year-olds, and will set an 18+ minimum for AI companion chatbots that simulate romantic or sexual relationships (GOV.UK).
The Case For It Is Real
The strongest version of this policy is not hard to state. Nine in ten parents who responded to the government's consultation — which drew more than 116,000 submissions — backed a ban, and two-thirds of young people surveyed agreed under-16s shouldn't be on certain platforms (GOV.UK). Recommender systems optimized for engagement do serve children content their parents never chose, and stranger-contact and livestreaming features carry documented grooming risk. A government facing that evidence and that public mandate has a legitimate case for acting, and defaulting the riskiest features off for 16- and 17-year-olds is a genuinely proportionate, low-cost intervention that doesn't require guessing anyone's exact age.
A Regulator Still Mid-Build
The under-16 ban does not arrive in a vacuum — it stacks on top of enforcement machinery Ofcom is still switching on. Ofcom's Protection of Children Codes, finalized under the existing Online Safety Act, already impose more than 40 measures on platforms and require "highly effective age assurance" for the riskiest services; Ofcom has already fined at least one adult-content provider £630,000 for age-check failures and can seek a court order to block a service in the UK altogether (Ofcom). Ofcom is due to publish its first statutory report on how effective that existing age-assurance regime has actually been in July 2026 — a month before ministers finalize the under-16 ban's own regulations. Legislating the harder, blunter instrument before the report on the softer one lands is a sequencing choice the government hasn't fully justified.
Australia's Head Start Is a Warning, Not a Blueprint
Britain has explicitly modeled its approach on Australia, which became the first country to enforce a social media minimum age on December 10, 2025. The early results are not encouraging for anyone expecting a ban to simply work. A survey of 1,050 Australians aged 12-15, conducted in March 2026 by the Molly Rose Foundation, found that more than 60% of teens who had accounts before the ban still had access to at least one platform, with TikTok, YouTube and Instagram each retaining more than half their under-16 users; roughly two-thirds of young users said the platforms had taken no action at all on their accounts (Fortune). Tellingly, VPNs weren't the main workaround — teens used fake birthdates, borrowed a parent's login, or wore printed masks to fool facial-age-estimation tools. That points to a structural problem rather than a fixable loophole: age-gating an entire product category invites evasion at the account-creation step in a way that narrower, harder-to-fake functionality restrictions do not.
The Verification Layer Has Its Own Costs
Ofcom's own guidance rules out simple self-declaration and steers platforms toward facial age estimation, ID or credit-card checks, or bank-verified digital identity. The Electronic Frontier Foundation's critique of the UK plan is worth taking seriously rather than dismissing as reflexive: there is no age-verification method that is simultaneously reliable, privacy-preserving and universally available, and a mandate applied to every UK user of a major platform — not just the under-16s the policy targets — pushes adults into submitting ID or biometric data to third-party verification vendors as a condition of ordinary internet use (EFF). Excluding WhatsApp and Signal from the ban, meanwhile, creates an obvious migration path: displaced teens move into encrypted messaging with none of the (imperfect) moderation and reporting infrastructure the banned platforms have built.
What Proportionate Actually Looks Like
The functionality restrictions — stranger-contact and livestreaming off by default, an 18+ floor on romantic AI companions — are the part of this package that survives scrutiny: targeted at specific, well-evidenced harms, applied without requiring a birth-certificate check on the entire user base. The blanket ban is the part built on Australia's model before Australia's model has demonstrated it works. With Ofcom's own effectiveness report landing in July 2026 and regulations not due before Parliament until December, ministers have a narrow window to let evidence catch up with ambition — narrowing the scope to functionality, or waiting on stronger verification standards, before entrenching an account-level ban that early data suggests a majority of motivated teenagers will simply route around.