Australia’s Teen Social Media Ban Still Lets Most Kids In

81.5% of Australian children under 16 were still using social media three months after the country barred them from holding accounts. Account-holding dropped, daily habits barely moved, and the regulator said weak age checks were the main reason kids kept or remade profiles.

That leak is now the export model. Meta’s U.S. teen time limits and Britain’s planned 2027 ban still have to tell a 15-year-old from a 16-year-old. On 8 September 2026, Canberra added an algorithm off-switch that does not close the gate.

Most Under-16s in Australia Never Left

The Social Media Minimum Age obligation took effect on 10 December 2025. It does not make it a crime for a child to open an app. It requires age-restricted platforms to take reasonable steps to stop Australians under 16 from having accounts.

eSafety’s 31 July 2026 evaluation, built with Stanford University’s Social Media Lab and following more than 4,000 children and families, is the first public look at what actually happened. Data collected in March show the share of under-16s with an account fell from 52.4% to 42.1%. Use of a restricted platform, with or without an account, fell from 85.9% to 81.5%. Daily use sat at 58%.

eSafety said most under-16s who already had accounts “were able to either retain them or create new ones at the three-month mark, with social media platforms’ failure to implement effective age assurance measures cited as the main reason.”

THE THREE-MONTH SCORE IN AUSTRALIA

Those are not rounding errors in opposite directions. Platforms told the regulator they had removed, deactivated or blocked about 4.7 million age-restricted accounts by mid-December 2025, then stopped more than 300,000 extra sign-ins by March 2026. Kids still used the products, including by watching public posts without logging in, which the law still allows.

Parental awareness of what girls and children aged 10 to 12 were doing declined in some groups. Most families told researchers the rules had not yet changed their lives. A smaller group said it was harder to keep up with friends. Another slice reported less pressure to be online.

eSafety is investigating Facebook, Instagram, Snapchat, TikTok and YouTube against the “reasonable steps” test. Those files were still open at the end of July 2026. The legal bar is not “did any child get through.” It is whether each company took steps that were available, effective and proportionate. That is a slower fight than the political slogan.

Why Age Checks Keep Failing Teen Social Media Bans

Australia never named one age-check method every platform had to use. Companies may infer age from activity, estimate it from a face, or accept a government ID. They are not allowed to force every Australian onto a national identity document. In practice, a lot of children still get through by typing a birth year that makes them 16.

A ban that cannot tell 15 from 16 is a posting rule, not a lock. Time limits, night blocks and chronological feeds only attach to accounts the company has already labelled as teen. If the label is wrong, the rest of the toolkit never fires.

The European Commission has been building the missing piece as a public utility rather than a per-app guess. Its EU age verification solution, a “mini wallet” that became feature-ready on 15 April 2026, is designed so a user can prove an age threshold without handing over a full identity file. The same stack is meant to sit inside national EU Digital Identity Wallets, which member states are supposed to offer by the end of 2026, and it can be set to 13+, 16+ or 18+.

That design still has a hard edge. To know that someone is under 16, a platform has to check the people who are not. False dates of birth, a sibling’s ID and a borrowed face already circulate as the teenage workaround culture. You do not age-gate children in isolation. You age-gate the adult internet and hope the kids do not walk around it. Australia has already shown they do.

The Poll Behind the Global Ban Wave

The policy is still popular where it has not been stress-tested. Pew Research Center surveyed 9,750 U.S. adults from 26 May to 1 June 2026 and published the results on 1 July 2026.

WHAT AMERICAN ADULTS TOLD PEW

• The headline ban: 56% of U.S. adults support blocking anyone under 16 from social media sites, 21% oppose it, and 23% are not sure.
• Parents: Support rises to 65% among adults with a child under 18, with 17% opposed.
• Softer tools: 85% back parental consent for a minor’s account (81% in 2023), 78% back age verification (71%), and 78% back time limits for minors (69%).

Republicans and Democrats both back a ban more than they oppose one, at 59% and 54%. Adults aged 30 to 49 are the most in favour, at 63%. The 18-to-29 group is the least enthusiastic and still nets support, at 52% in favour and 26% opposed.

That is a mandate for action, not a mandate for a method. The same adults who want a ban also want age checks, and Australia is the proof that wanting the check is easier than running one at national scale.

WhatsApp, Discord and Roblox Stay Open

The Australian law carves out services whose main job is messaging, calling, gaming, health or classroom work. Social sharing has been moving into those rooms for years. A feed ban that leaves group chats untouched invites the social graph to follow the exemption.

PLATFORMS INSIDE AND OUTSIDE THE AUSTRALIAN RULE

• In scope: Facebook, Instagram, Kick, Reddit, Snapchat, Threads, TikTok, Twitch, X and YouTube must take reasonable steps to keep under-16s from holding accounts.
• Out of scope: Discord, GitHub, Google Classroom, LEGO Play, Messenger, Pinterest, Roblox, Steam (including Steam Chat), WhatsApp and YouTube Kids sit outside the account ban.
• Logged out is legal: Under-16s may still watch public posts that do not require an account, which is how YouTube in particular remains a daily habit without a profile.
• Meta’s split: Instagram and Facebook are in. Messenger is out. A teen locked out of Reels can still live in the same company’s chat product.

On 8 September 2026 the government tried to close that gap from another angle, by extending a duty of care to messaging apps, games and AI chatbots rather than by adding them to the account ban. The original hole remains. A 14-year-old who cannot (in theory) hold TikTok can still hold WhatsApp, Discord and Roblox, and can still watch YouTube without signing in.

Two Hours a Day, Once the App Knows Your Age

On 26 August 2026, New York Attorney General Letitia James and a coalition of 50 other attorneys general announced a settlement with Meta over claims it built Facebook and Instagram to keep young people hooked. The money is up to $17.1 billion, with at least $12.1 billion guaranteed. New York’s share is at least $819 million and up to $1.15 billion, earmarked for mental-health and school programs. New Mexico had already won $375 million plus $567 million in its own case, $942 million in total, and stayed out of the coalition deal.

Children in New York and nationwide are suffering while companies like Meta reap immense profits by intentionally addicting them to their social media platforms. With significant new resources for our communities and comprehensive restrictions on Meta’s platforms for young people, we are taking a major step towards breaking the cycle of social media addiction.

Letitia James, New York Attorney General, 26 August 2026 settlement announcement

The product terms are the part other governments will copy. They only apply to people Meta has identified as under 18.

DEFAULT RULES FOR U.S. USERS UNDER 18

The base restrictions last at least five years. Meta’s chief legal officer, C.J. Mahoney, said the time limits, night mode and school-hour mute “set the right path forward for our whole industry, but this framework will only work if all our peers join us.” He is talking about competitive fairness. The prior problem is identification. A two-hour clock cannot run on a 15-year-old who registered as 21, which is the failure mode Australia has already measured in public.

Draft Laws Pull Chatbots and Messaging In

Prime Minister Anthony Albanese and Communications Minister Anika Wells released the My Feed, My Way draft laws on 8 September 2026. Platforms would have to ask new and existing users whether their default feed should be algorithmic or limited to people and creators they chose to follow. That choice is for Australians over 16, the cohort already allowed to hold accounts.

The same draft, branded a Digital Duty of Care, reaches past the original ban. Online games, apps and AI chatbots would have to protect under-18s from addictive design and from content that promotes eating disorders, hostile ideas about women, pornography, crime-glorifying stunts, or serious bullying. Failure could cost up to $109.2 million. eSafety would also get faster removal powers for “nudify” apps.

The duty of care goes beyond social media to ensure children are protected from harmful content and features, including on chatbots and gaming platforms, apps and messaging platforms.

Anika Wells, Minister for Communications, 8 September 2026

Wells is describing the migration path the first law left open. In June 2026 the government had already doubled the maximum penalty for systematic failures of the account ban to A$99 million from A$49.5 million. More money does not invent a working age signal. It raises the price of getting caught without one.

HOW THE RULES STACKED UP

1. 10 December 2025: Australia’s social media minimum age takes effect for under-16 accounts.
2. 16 January 2026: eSafety says platforms have restricted about 4.7 million age-restricted accounts.
3. March 2026: A compliance update finds more than 300,000 extra sign-ins blocked and opens investigations into five major platforms.
4. 31 July 2026: The three-month family study is published; 81.5% of under-16s are still using restricted platforms.
5. 26 August 2026: Meta agrees U.S. teen caps, a night block and a school mute that all depend on age labels.
6. 8 September 2026: Canberra releases duty-of-care draft rules for algorithms, chatbots, games and messaging, with a compliance update on the account ban due the following week.

The government wants comments from platforms, civil-society groups and advocates, and says a bill will go to Parliament later in 2026. Until a uniform age check exists, the new duties attach to the same uncertain teen cohort as the old ones.

Britain Plans a 2027 Copy of the Same Model

The UK government said on 15 June 2026 that it will ban social media for under-16s, using Australia’s model, with the first protections expected in Spring 2027. A national consultation found 9 in 10 parents in favour. The planned list includes Snapchat, TikTok, YouTube, Instagram, Facebook and X. Messaging services such as WhatsApp and Signal are not meant to be in the ban. On 15 July 2026, ministers added a default midnight-to-6 a.m. curfew and a switch-off of autoplay and personalised feeds for 16- and 17-year-olds, plus extra blocks on livestreaming and stranger contact.

That package is Australia plus Meta’s night mode, exported before Australia has shown it can keep 15-year-olds off the original apps. Britain will still need an age signal that works on the first login. The EU wallet is being built for that job. Australia declined to mandate any one method and then watched most under-16s keep using the products.

Wells said a new social media minimum age compliance update would come the week after 8 September 2026. The investigations into Facebook, Instagram, Snapchat, TikTok and YouTube are the test of whether “reasonable steps” means anything once a child can still type a fake birthday.