On June 10, 2026, Canada tabled Bill C-34, the Safe Social Media Act, which would enact a new Digital Safety Act, stand up a Digital Safety Commission of Canada, and — at its centre — prohibit social media accounts for anyone under 16, with an exemption pathway for platforms that can demonstrate sufficient safeguards. The bill, now at second reading in the House of Commons, also imposes duties on AI chatbots, including suicide-intervention requirements. Health Minister Marjorie Michel framed the rationale in developmental terms: platforms and chatbots "are designed to capture attention" and "do not support healthy childhood development."
The template is unmistakably Australian. And that is precisely why the rest of the world should look closely at what Australia's experiment has — and has not — yet demonstrated.
The numbers that built the template
Australia's Online Safety Amendment (Social Media Minimum Age) Act 2024 received assent on December 10, 2024, and its core obligation took effect a year later, on December 10, 2025. Ten platforms — Facebook, Instagram, TikTok, YouTube, Snapchat, X, Reddit, Threads, Twitch and Kick — were assessed as age-restricted and required to take reasonable steps to prevent under-16s from holding accounts, on pain of fines of up to A$49.5 million for systemic non-compliance.
The headline result came fast. In a January 22, 2026 media release, the eSafety Commissioner reported that platforms had restricted access to about 4.7 million under-16 accounts. Meta alone removed roughly 550,000 accounts by the day after commencement. Commissioner Julie Inman Grant called the figures "encouraging" against a population of roughly 2.5 million Australians aged 8–15 — and against pre-ban estimates that 84% of 8–12-year-olds held social media accounts despite existing platform age rules.
Those deactivation counts are the export product. France's National Assembly approved an under-15 ban in January 2026. Denmark struck a political agreement in November 2025 setting its cut-off at 15, with parental authorisation possible from 13. Greek Prime Minister Kyriakos Mitsotakis announced on April 8 that Greece will bar under-15s from January 1, 2027. Poland's governing party said in February it is preparing similar legislation. The European Commission, unveiling a privacy-preserving age-verification app on April 15, declared there were "no more excuses." Canada is the first country outside Europe to legislate the full package.
The strongest case for the ban
Be fair to the regulators: the case for a minimum age is not hysteria. Engagement-optimised design genuinely exploits developmental vulnerabilities, and individual parents face a collective-action problem — opting one child out of social media means opting them out of their peer group, which is why voluntary parental controls have chronically underperformed. Australia's 4.7 million deactivations also answered the most common pre-ban objection: that enforcement was technically impossible. It plainly is possible, at least at the level of account removal. Those are real achievements, and the sponsors of Bill C-34 are not wrong to take them seriously.
What the template actually exports
The problem is what the deactivation counts don't measure. Six months in, there is no evidence yet on the outcomes the law was passed to improve — adolescent mental health, anxiety, social connection. Account removals are a compliance metric, not a wellbeing metric, and Australia's own evaluation of the policy's real-world effects has barely begun. Countries are copying the answer before the experiment has reported.
Compliance itself is shakier than the headline suggests. eSafety's first Social Media Minimum Age compliance report, published in March 2026, detailed significant concerns about Facebook, Instagram, Snapchat, TikTok and YouTube, with children continuing to create new accounts and pass age-assurance checks; the regulator has publicly signalled it is weighing court action against several platforms. A ban that the most-used platforms cannot reliably enforce against motivated 14-year-olds is a ban that pushes those users toward smaller, less moderated services that face no scrutiny at all.
Then there is the cost the template imposes on everyone else. As University of Ottawa law professor Michael Geist notes of C-34, identifying under-16s requires age-checking the entire user base — a privacy intrusion on tens of millions of adults, however "temporary" the verification is claimed to be. Geist counts roughly 50 critical decisions that C-34 defers to future cabinet regulations and a commission that does not yet exist, and warns that its AI chatbot regime uses a capability test that could sweep in every general-purpose assistant. Canada has, in other words, copied Australia's blunt instrument and attached an open-ended regulatory mandate to it.
Evaluation before emulation
A proportionate version of this agenda exists. The European Commission's age-verification app, built on zero-knowledge proofs that confirm age without exposing identity, is a meaningfully better architecture than per-platform document uploads, and binding safety-by-design duties target the actual mechanism of harm — engagement optimisation — rather than access itself. Any jurisdiction adopting age limits should pair them with mandatory independent evaluation and sunset clauses tied to results.
Australia earned the right to run a world-first experiment, and it deserves credit for publishing compliance data candidly, warts and all. But an experiment is only useful if you wait for the results. Five countries legislating the template inside six months — on deactivation counts alone — is not evidence-based policymaking. It is policy contagion, and the open internet will bear the costs long before anyone can say whether children are better off.