Six weeks is not long enough to know whether a law worked. It is long enough to know whether it did anything at all, and Australia's eSafety Commissioner published the first real numbers on January 16, 2026 — a month after the ban itself took effect. I went through the regulator's own releases and the March 2026 rule amendments rather than the initial headlines, because "world-first ban" stories tend to get less careful the further they travel from Canberra.
What the law actually does
The Online Safety Amendment (Social Media Minimum Age) Act requires designated "age-restricted social media platforms" to take reasonable steps to prevent Australians under 16 from holding accounts. It took effect December 10, 2025. Crucially, the list of covered platforms wasn't fixed at launch: the Online Safety (Age-Restricted Social Media Platforms) Amendment Rules 2026, made in March 2026, refined the eligibility criteria — a service qualifies if its sole or significant purpose is to enable social interaction between users, users can link to or interact with each other, and users can post material. As of the March 2026 update, the covered list runs: Facebook, Instagram, Kick, Reddit, Snapchat, Threads, TikTok, Twitch, X, and YouTube. Non-compliant platforms face a civil penalty of up to 150,000 penalty units — cited as roughly A$54.6 million in some coverage and roughly A$49.5 million in others, depending on the per-unit value applied.
The first numbers
On January 16, 2026, eSafety reported that age-restricted platforms had removed access to 4.7 million under-16 accounts across Australia by mid-December 2025 — within about a month of the law taking effect. That figure came in far above pre-launch estimates and works out to more than two account-closures for every Australian aged 10 to 16, based on population data (closures, not unique users, since one teenager blocked on three platforms counts three times). Meta alone removed roughly 550,000 underage accounts across Instagram, Facebook, and Threads. eSafety Commissioner Julie Inman Grant was notably unwilling to declare victory: she acknowledged that some underage accounts remain active on the platforms and that it was still too early to call the rollout fully compliant.
What isn't measured yet
Two things the regulator has not published, and that no credible source has filled in independently: a circumvention rate (how many of the 4.7 million closed accounts have already been recreated via VPNs, falsified birthdates, or borrowed adult identities), and any wellbeing or mental-health outcome data, which would need far more than six weeks to be meaningful. Coverage that reports either of those as settled numbers is getting ahead of what eSafety itself has released.
The rest of the world is watching the data, not the headline
What makes this more than an Australian domestic story is how quickly other governments cited Australia's own numbers while drafting their own versions:
| Jurisdiction | Status as of mid-2026 |
|---|---|
| Malaysia | Announced a ban on under-16 social media access starting in 2026 |
| New Zealand | Legislation modeled on Australia's law, with a parliamentary committee report due in early 2026 |
| Denmark | Announced a ban for under-15s, with a parental-consent carve-out down to age 13 |
| Norway | Government committed to introducing a bill by end of 2026 banning under-16 social media use, with tech companies responsible for age verification |
| Poland | Ruling party drafting legislation to ban under-15 access and mandate age verification |
| France | Considering a under-15 ban plus a 10pm-8am platform curfew for 15-18 year-olds |
| United Kingdom | Reportedly considering an Australia-style under-16 ban alongside new AI chatbot safety rules |
| European Union | European Parliament called for a similar under-16 ban in November 2025; European Commission President Ursula von der Leyen has said she is studying Australia's results directly |
The honest read
Australia's law is the first real-world test of platform-side age enforcement at national scale, and the eSafety data confirms platforms can, in fact, close millions of accounts on a government deadline when the penalty is credible. What it does not yet confirm — and what every government in the table above is really waiting to find out — is whether closed accounts stay closed, and whether any of this moves the outcomes the law was written for. Both questions need another year of data, not another six weeks.
Sources: eSafety Commissioner media release, Jan 16 2026, eSafety social media age restrictions page, Online Safety (Age-Restricted Social Media Platforms) Amendment Rules 2026, Federal Register of Legislation, eSafety March 2026 compliance update PDF, NBC News, The Conversation, ScienceAlert on copycat countries.
