The Social Media Ban That Failed at Home and Conquered the World

In February, Anthony Albanese stood before reporters and named the countries following Australia's lead. Spain. France. Denmark. Indonesia. Malaysia. The United Kingdom. His government had removed 4.7 million child accounts from social media platforms, he said, and the world was watching. ^[1]

Our ban came into force on December 10. It is working across 10 platforms. The feedback from parents is so positive, and the feedback from young people as well. — Anthony Albanese

Six months later, the evidence from Australia itself is in, and it is devastating. A BMJ study found that 85 percent of Australian 12-to-15-year-olds were still accessing the restricted platforms. A separate survey by the Molly Rose Foundation found that seventy percent of those teens said circumvention was easy, and eSafety Commission briefings showed no meaningful drop in TikTok or Instagram downloads. ^[2][3][4] The pioneer is exporting a product that is failing at home, and it is doing so with confidence. That gap — between the model's evidentiary record and its political momentum — is the story of youth social media regulation in 2026. The playbook is now recognizable across four continents: age verification at the app-store or operating-system level, bans on addictive design features like infinite scroll and autoplay, and platform liability for harm to minors. It appears in Connecticut's proposed legislation, in North Carolina's and Michigan's bills, in Canada's Senate deliberations, in the Philippines' Senate Bill 601, and in the EU's newly announced bloc-wide minimum age. ^[5][6][7][8] What varies is not the tool but the language used to justify it. Emmanuel Macron framed France's ban in explicitly sovereign terms.

Because our children's brains are not for sale—neither to American platforms nor to Chinese networks. — Emmanuel Macron

The European Commission, by contrast, is building its age-verification architecture — an open-source app, harmonized across the bloc — through product-safety law. Ursula von der Leyen has cast algorithmic harm to minors as a design defect, shifting the burden of proof onto the platforms. ^[8]

This is not about whether children can access social media. It is about when social media can access our children. — Ursula von der Leyen

Pakistan's Senate formed a multi-ministry committee citing Australia's model directly. The Philippines introduced its bill in the language of child safety and national security. India's Goa state is studying the Australian Online Safety Amendment Act. The same mechanism — keep children off platforms, make platforms liable for what reaches them — travels across regime types without friction. ^[9][7][10] The United Kingdom is the sharpest demonstration that political logic, not evidence, is driving the diffusion. In March, Parliament voted 307 to 173 against a blanket under-16 ban, opting instead for flexible regulatory powers. By June, Keir Starmer had announced the Wellbeing and Schools Act 2026, imposing precisely the ban Parliament had rejected, to take effect in spring 2027. ^[11][2]

social media was making children unhappy — Keir Starmer

The reversal did not rest on new evidence. It rested on executive momentum and public consultation — the political imperative to be seen acting against addictive algorithms. Even the Molly Rose Foundation, the advocacy group that found roughly two-thirds of Australian teens still accessing restricted platforms, warned the UK against copying the model, recommending regulators target addictive design and business models rather than impose access bans. ^[3] And then there is the hinge. Australia itself is pivoting. The same government that exported the access-ban model is now moving toward a "digital duty of care" — a liability framework that obliges platforms to protect users from foreseeable harm, rather than attempting to keep minors off platforms entirely. The BMJ study and the Molly Rose Foundation survey, which together showed between 61 and 85 percent of teens bypassing the ban, made the access model untenable at home even as it was being adopted abroad. ^[2]

I don't have potent powers. — Julie Inman Grant

The countries copying the ban are importing a model its inventor is already leaving behind. Beneath the policy diffusion is a structural layer that will outlast any single ban. The age-verification checkpoints being built to enforce teen restrictions — app-store identity gates, government-issued credential checks — are the same infrastructure governments want for broader digital identification. Spain's prime minister has pledged an end to online anonymity. The United Kingdom is mandating digital IDs for employment. The United States is tying REAL ID to travel. The EU's open-source age-verification app, presented as a child-safety measure, is a general-purpose identity checkpoint. ^[12][8] Meta has backed app-store-level age verification across multiple jurisdictions, arguing it avoids a patchwork of inconsistent national rules. The position is strategically legible: it shifts the compliance burden and liability onto Apple and Google, the operating-system gatekeepers, while positioning Meta as a cooperative partner of the regulatory project. ^[6][13]

This is the only way to guarantee consistent, industry-wide protections for young people, no matter which apps they use, and to avoid the whack-a-mole effect of catching up with new apps that teens will migrate to in order to circumvent the social media ban law. — Meta

The same platform-liability mechanism that enforces youth bans is being deployed for very different speech targets. Brazil's Lula signed decrees holding platforms liable for illegal content, mirroring the EU's Digital Services Act. Pakistan warned platforms to remove militant content and proposed Brazil-style fines. Russia banned Human Rights Watch, designated Alexei Navalny's foundation a terrorist group, and threatened to block WhatsApp entirely — all through the same compliance language that, in other jurisdictions, is sold as child protection. ^[14][15][16] The legal counter-pressure is mounting. Reddit has filed a High Court of Australia challenge arguing the under-16 ban burdens political communication. Two Australian fifteen-year-olds, backed by the Digital Freedom Project, have brought a separate suit. In the United States, NetChoice has successfully enjoined state-level restrictions on First Amendment grounds, and Virginia's one-hour daily limit is on appeal. ^[17][11] These challenges may eventually force the evidence question that the political logic has deferred. But the infrastructure being built to enforce the bans — the identity gates, the liability frameworks, the compliance apparatus — will not be dismantled simply because one policy justification weakens. Youth safety was the politically saleable wedge. The architecture it justified is already in place, and it is designed for uses far beyond keeping a fifteen-year-old off TikTok.