Age verification online has moved from a debate to an implementation problem, and the countries that went first have produced some genuinely useful evidence. Australia has banned social media for under-16s outright. The European Commission is drafting an EU-wide age verification requirement. And in a few European countries the strictest version has been running for years in specific sectors, with identity confirmed against a national digital ID before access rather than after: in Lithuania, nobody can reach online slot machines Lithuania licenses without verifying who they are first, and the minimum age there is 21 rather than 18.
The Lithuanian model is instructive because it is not a warning screen. A licensed operator such as Twinsbet has to confirm identity before a first play, a first deposit and a first withdrawal, all three, using either a national eID service or an uploaded document with a selfie. Users must set deposit and session-time limits before they start. Reducing a limit takes effect at once, raising one takes 48 hours. Anonymous access is not permitted at all.
Australia proved enforcement is possible
The ban on social media accounts for under-16s took effect on 10 December 2025. Most commentary assumed platforms would ignore it. By 16 January, 4.7 million accounts had been removed or restricted, with Meta reporting around 550,000 of its own.
Covered services include Facebook, Instagram, Kick, Reddit, Snapchat, Threads, TikTok, X, YouTube and Twitch. Messaging apps are excluded, which is a significant carve-out given how much teenage communication has already migrated there. The maximum penalty is 49.5 million Australian dollars.
Whether the ban improves teenage wellbeing is a separate question and will take years to answer. What it settled quickly is that “unenforceable” was not the right objection.
Brussels is going after the interface, not the content
In July 2026 the European Commission issued preliminary findings against Meta under the Digital Services Act. The complaint targets infinite scroll, autoplay, push notifications and hyper-personalised feeds as manipulative design under articles 25, 34 and 35. Remedies under discussion would make endless feeds and autoplay opt-in and require built-in break prompts. The theoretical maximum fine is 6 per cent of global turnover.
Preliminary findings against TikTok were issued in February. A Digital Fairness Act announced on 12 May 2026 would generalise the approach, restricting addictive design across services and mandating a pan-European age verification app.
The evidence on screen time is stronger than it was
For years the research on screen time was correlational and therefore easy to dismiss. Two controlled trials published in 2025 changed that.
The first, in PNAS Nexus, blocked mobile internet on 467 participants’ phones for two weeks. Daily screen time fell from 314 minutes to 161. Mental health improved with an effect size the authors noted exceeded the meta-analytic effect of antidepressants. The second, in BMC Medicine, capped 111 students at two hours a day for three weeks and recorded a 27 per cent fall in depression scores, rising to 40 per cent among those who actually kept to the limit.
But the panic is overstated
Pew Research surveyed 1,458 US teenagers in autumn 2025. Only around 30 per cent of TikTok users thought they spent too much time on the app. More than 60 per cent considered their usage about right and roughly 70 per cent described their experience as mostly positive, against 3 per cent who called it negative.
Both findings can be true. Reducing use helps people who reduce it. That does not mean most teenagers are in crisis.
What it means here
For parents, the practical takeaway is that limits set in advance work considerably better than limits imposed after a problem appears, and that a delay on loosening them is the single most effective design feature in any of these systems. That principle is available on most phones today without waiting for legislation.
