Australia’s Under-16 Social Media Ban Shows Little Early Impact

Three months after Australia became the first country to ban under-16s from social media, 85 per cent of adolescents covered by the law were still using restricted platforms, mostly through their own accounts, according to a peer-reviewed study in The BMJ. The independent evaluation, carried out by the University of Newcastle and accepted for publication on 12 June 2026, found “insufficient evidence of any substantive early effects” of the Social Media Minimum Age Act. The UK government announced its own copy of the policy on 15 June 2026, promising stronger age verification and an extension into gaming and livestreaming that Australia did not attempt.

Prime Minister Keir Starmer used a Downing Street speech to confirm the ban would cover Snapchat, TikTok, YouTube, Instagram, Facebook and X, with a target of spring 2027 for implementation. The package adds restrictions on under-16s livestreaming or communicating with strangers on gaming platforms, a category that includes Roblox. A UK government spokesperson told the Guardian that the British approach “goes further than the Australia model and will be underpinned by stronger, more effective age verification checks”. Technology Secretary Liz Kendall has separately acknowledged the ban will not be a “complete silver bullet”. The BMJ study, on its own terms, asks whether that louder promise changes the outcome.

What the Newcastle Researchers Actually Measured

The study followed 408 Australian adolescents aged 12 to 17, drawn from a baseline of 436, immediately before and approximately three months after the legislation came into force in December 2025. The team used a sharp regression discontinuity design, comparing outcomes for those just under 16 with those just over 16, an approach the authors describe as appropriate for a policy-level intervention where randomisation is not feasible. The full paper is published as the Newcastle University assessment of the early effects.

• More than 85% of participants under 16 reported still using social media platforms subject to the Act at follow-up.
• Most of those who kept using platforms did so on their own accounts (54-68% across age bands).
• Two-thirds of those still using restricted platforms reported encountering some form of age verification, most commonly self-declared age (24-39%) or a selfie upload (13-27%).
• 15-19% of under-16s reported using a “fake” account to access platforms.
• Between 6% and 11% reported accessing platforms through a private browser.

Daily use held almost steady for the youngest cohort. Among 14 to 15-year-olds it dipped from 78% to 69%. Among those aged 16 and over, who were not legally subject to the ban, it rose from 80% to 89%. Time spent on social media per day, measured on an ordinal scale, fell from 3.40 to 3.13 for the 14-15 group and was stable for the other bands.

On the two co-primary outcomes, daily use and time spent, the regression discontinuity found no statistical discontinuity at the age-16 threshold (P≥0.60). In plain terms, the data did not show a clean break between those inside the law’s reach and those just outside it.

Lead investigator Dr Courtney Barnes, a postdoctoral research fellow at Newcastle, framed the finding in a statement issued through the university: “In this study, we observed very little change in adolescent social media use in the three months following the introduction of age restrictions, based on self-reported data collected before and after implementation.” Co-investigator Professor Luke Wolfenden, a behavioural scientist at the same institution, said the results point to “the importance of how age assurance is operationalised in practice”.

How the Ban Got Around So Easily

The most uncomfortable number in the study is also the simplest. Most adolescents who kept using restricted platforms did so on their own existing accounts. The primary mechanism the Australian law set in motion was supposed to be the deactivation of those accounts by the platforms, and most of them kept working anyway.

Where age verification was encountered, it was usually soft. Self-declared age, the child typing a number into a box, was the most common check, reported by 24% to 39% of under-16 users across age bands. A selfie upload was next at 13% to 27%. Only 5% of 12 to 13-year-olds and 11% of 14 to 15-year-olds, according to the Guardian’s reading of the study, were asked for a photo of official ID. A system designed to be porous enough to keep minors off the most harmful design features was, in practice, porous enough to let most minors through.

Active circumvention was less common than passive slippage but still measurable. 15 to 19 per cent of under-16s reported using a fake account. Between 9% and 29% accessed platforms through someone else’s account. The BMJ paper’s own language is striking: “limited implementation, incomplete compliance, and substantial circumvention of social media restrictions.” Earlier polling by the Molly Rose Foundation, published in March 2026, found 61% of Australian 12 to 15-year-olds who had accounts before the ban still had access to at least one restricted platform, with seven in ten of those who kept using restricted sites telling pollsters it was “easy” to get around the rules.

The UK’s Bigger Bet on the Same Idea

Starmer’s announcement, made in the Prime Minister’s full Downing Street remarks on the ban, frames the under-16 ban as a personal and political wager. “I am not prepared to compromise on the safety and happiness of our children,” he said, before conceding that “some technology companies want us to think that social media is unchangeable, part of an almost natural order”. The law will exclude messaging services such as WhatsApp and Signal. Starmer said the government would use powers from the Children’s Wellbeing and Schools Act to “adapt as technology changes”.

The political argument for the UK doing more than Australia rests on age verification. A UK government spokesperson said the British system would use “highly effective age-verification measures” stronger than the Australian checks the BMJ paper describes. The government also says its package extends beyond the six social platforms into gaming and livestreaming, areas Australia left alone. Under-16s will be unable to livestream, and unable to communicate with strangers on gaming platforms including spring 2027 targets such as Roblox.

The moral framing was set out the same week by Kanishka Narayan, a UK government minister, in Kanishka Narayan’s post on the policy: “Our children’s safety is never an economic calculation. It is a moral priority for this Government.” It is a stance the campaign groups who have pushed hardest for child online safety do not share.

The Voices Saying a Ban Cannot Do This Alone

Ian Russell, the chair of the Molly Rose Foundation and the father of 14-year-old Molly Russell, who took her own life in 2017 after viewing harmful content online, published a direct reply to the Prime Minister in the Guardian. In Ian Russell’s open letter to the Prime Minister, he argued that the UK was choosing “a politically easy route which the evidence shows, and experts warn, will not work”. He cited the Australian finding that 60% of teens were still accessing social media after the ban and warned that the same outcome would follow in Britain. He also pointed to research his own foundation had shared with Downing Street showing that one in two girls aged 13 to 17 still see high-risk suicide, self-harm and body-image content online, most of it algorithmically driven.

Andy Burrows, the chief executive of the Molly Rose Foundation in the UK, was blunter still. In a statement carried in the BMJ findings and expert reactions collected by the Guardian, he said: “Unless ministers have a coherent plan to urgently learn lessons, the UK’s ban will similarly unravel. Parents will be left with false hope and a misplaced sense of their children’s safety. The next prime minister must enter Downing Street with a convincing strategy that properly protects children from online harm, rather than relying on a performative ban which, as this research suggests, is unlikely to improve our teens’ mental health and wellbeing.”

Unless ministers have a coherent plan to urgently learn lessons, the UK’s ban will similarly unravel. Parents will be left with false hope and a misplaced sense of their children’s safety.

Rachel de Souza, the Children’s Commissioner for England, said the ban “should not be seen as a silver bullet”. Her objection is structural: harmful features such as personalised algorithms operate across all online services, not just the platforms a ban can name. “We have to go further,” she said, “so that all online services, not just social media platforms, that use harmful features and functionalities should be banned from access to all children, not just those under 16.”

Professor Dennis Ougrin, a consultant child and adolescent psychiatrist at Queen Mary University of London, read the BMJ paper more cautiously. The study should serve as “an important early reality check for policymakers,” he said, but it is “too early to conclude that the policy has failed”. The question that matters, he argued, is not whether use falls but whether restrictions improve the outcomes that prompted the legislation in the first place: mental health, sleep, exposure to harmful content and self-harm. The Newcastle data, on his reading, cannot answer that question yet.

Where Australia Goes, Where the UK Goes

Australia’s Online Safety Amendment (Social Media Minimum Age) Act 2024 came into force on 10 December 2025. The eSafety Commissioner reported that, by mid-December 2025, age-restricted platforms had removed access to 4.7 million under-16 accounts. The legislation covers 10 designated platforms: TikTok, Instagram, Snapchat, YouTube, X, Facebook, Reddit, Twitch, Threads and Kick. Messaging services are excluded, in line with the UK approach.

Outside the BMJ cohort, the picture from March 2026 polling was consistent with the academic finding. The Molly Rose Foundation found that 61 per cent of Australian 12 to 15-year-olds who had pre-ban accounts still had access to at least one restricted platform. YouthInsight polling of 1,050 Australians aged 12 to 15 found that seven in ten children still using restricted sites said it was “easy” to get around the ban. For related reading on how a similar debate is unfolding across the Tasman, see the New Zealand teen social media reckoning.

Side by side, the two policies share an age threshold and a list of social platforms but diverge on what sits beyond them.

What Three Months Cannot Settle

The Newcastle team is careful about what the data can and cannot say. The 408 adolescents are drawn from a single Australian state, the outcomes are self-reported, and the study measures use rather than harm. A 12-month follow-up is planned for December 2026, a window that will pick up whether platforms tighten their checks, whether the gaming and messaging extensions change behaviour, and whether the figures on daily use shift further. For context on where the policy’s blind spots may sit in the meantime, see where the Australian ban may miss the youngest risk.

Wolfenden’s own summary of the work is the line the BMJ paper ends on: “It takes time for policy implementation to stabilise and for impacts, particularly on wellbeing, to emerge.” The UK has committed to its version of the same experiment. The Australian data does not say the policy cannot work. It says that, at three months, with the age-assurance system that was actually built, it largely did not.

Frequently Asked Questions

What does the BMJ study actually find about Australia’s under-16 ban?

More than 85% of under-16s in the cohort were still using restricted social media platforms three months in, most often on their own accounts. Daily use dropped slightly for 14 to 15-year-olds, from 78% to 69%, held steady for 12 to 13-year-olds, and rose for those just over 16, from 80% to 89%. The authors found no statistical discontinuity in either primary outcome at the age-16 threshold.

Which platforms are covered by each country’s ban?

Australia’s law covers 10 designated platforms: TikTok, Instagram, Snapchat, YouTube, X, Facebook, Reddit, Twitch, Threads and Kick. The UK plan covers six social platforms, Snapchat, TikTok, YouTube, Instagram, Facebook and X, and adds restrictions on under-16s livestreaming and on contacting strangers on gaming platforms such as Roblox.

How are Australian children getting around the rules?

Mostly through their own accounts, which most platforms did not deactivate. Where age verification ran, the most common checks were self-declared age and a selfie upload. About 15% to 19% used a fake account, 6% to 11% used a private browser, and between 9% and 29% accessed platforms through someone else’s account.

What is the UK doing differently from Australia?

The UK government says it will use “stronger, more effective age verification checks” and has explicitly extended the policy to gaming and livestreaming, where Australia did not legislate. Ministers say the Children’s Wellbeing and Schools Act gives them powers to adapt the rules as technology changes.

What are the experts saying about the BMJ study?

Researchers call it an important early reality check for policymakers, but Professor Dennis Ougrin, a child and adolescent psychiatrist at Queen Mary University of London, has said it is “too early to conclude that the policy has failed”. Campaigners including the Molly Rose Foundation argue a ban alone cannot fix the underlying product-safety problems and risks giving parents false hope.