One benefit of going second is learning the practical lessons
In a world where childhood increasingly unfolds on screens, governments are beginning to ask whether the law can protect what parenting and culture have struggled to preserve. Britain's online safety minister traveled to Australia this May to study the first major national attempt to bar teenagers from social media — a law now six months old and already revealing the distance between legislative intent and lived reality. The journey was less a search for a blueprint than a reckoning with the enduring truth that rules, however well-crafted, must contend with human ingenuity, corporate interest, and the quiet persistence of young people finding their way around every wall built for them.
- Australia's under-16 social media ban, in force since December, has left two-thirds of its intended subjects still scrolling — a gap between law and compliance that no government can easily ignore.
- Five of the world's most powerful platforms are under formal investigation for failing to enforce the ban, with one allowing facial age verification attempts up to 24 times a day, reducing a safeguard to a lottery.
- The threat of fines reaching A$49.5 million per breach has prompted platforms to tighten their systems, but regulators and teenagers remain locked in a continuous game of adaptation and circumvention.
- Britain is preparing its own restrictions — potentially live by year's end — and is using Australia's stumbles as a map of the terrain, focusing on what 'robust age assurance' actually requires in practice.
- Beyond enforcement, Australia's ban has done something unexpected: it has sparked a national conversation among parents, carers, and youth workers that may prove as consequential as any regulatory mechanism.
Kanishka Narayan, Britain's online safety minister, spent a week in Australia this May on what amounted to a policy reconnaissance mission — studying a country that moved first on teenage social media restrictions and is now living with the consequences. What he found was a law that commands attention but struggles to command compliance: roughly two-thirds of under-16s remain on social media despite a ban that took effect in December.
The stakes for Britain are immediate. Within weeks, UK officials are expected to announce their own restrictions — age limits, design changes targeting addictive features, or both — potentially in place before the year is out. Narayan's visit included meetings with federal and state ministers, school students, and Australia's eSafety commissioner, Julie Inman Grant, who offered an unvarnished account of the enforcement challenge.
Five major platforms — Snapchat, TikTok, YouTube, Instagram, and Facebook — are under investigation for non-compliance. One platform allowed users to attempt facial age verification an average of 24 times a day, turning a security measure into something closer to trial and error. Since regulators signaled that breaches could carry fines of up to A$49.5 million, companies have introduced tighter checks and new verification steps — but the workarounds continue to evolve.
Narayan was candid about Britain's advantage in going second. The central challenge, he said, was not passing a law but making it work — identifying which platforms to restrict and building age verification systems that hold up under pressure. He described 'robust age assurance' not as a problem to be solved once, but as a continuous process of adjustment and regulatory vigilance.
What struck him most, though, was less measurable. Australia's ban had changed the cultural conversation — parents, youth workers, and families were engaging with questions about teenage life online in ways they hadn't before. That shift in awareness, Narayan suggested, might carry as much weight as any enforcement mechanism.
The two nations also signed a memorandum of understanding on AI security, agreeing to share intelligence on frontier models and cyber vulnerabilities. When asked whether tech companies might be trading early system access for regulatory goodwill, Narayan was direct: security, he said, could not be built on trust in companies alone.
The lesson Australia is teaching Britain is an old one in new clothes: legislation is the beginning of the work, not the end of it.
Kanishka Narayan, Britain's online safety minister, spent a week in Australia this May studying how a nation half a world away is trying to keep teenagers off social media. What he found was instructive but sobering: a law that looks good on paper but struggles in practice, with roughly two-thirds of under-16s still managing to stay online despite a ban that took effect in December.
The Australian experiment matters to Britain because the UK government is preparing its own crackdown. Within weeks, officials expect to announce restrictions that could include age limits on platforms, changes to features designed to hook young users, or both. If approved, these rules could be in place by year's end. Narayan's visit—which included meetings with state and federal ministers, school students, and Julie Inman Grant, Australia's eSafety commissioner—was essentially a reconnaissance mission: what works, what doesn't, and what pitfalls should Britain avoid.
The core problem Australia faces is enforcement. Five companies—Snapchat, TikTok, YouTube, Instagram, and Facebook—are under investigation for failing to comply with the ban. Inman Grant revealed to parliament that some platforms had been careless about their gatekeeping. One allowed users to attempt facial age verification up to 24 times a day on average, turning a security measure into something closer to a game of chance. Since regulators put these companies on notice that breaches could cost them up to A$49.5 million each, there have been improvements: more rigorous age checks, new verification steps when users try to change their birth date, and easier reporting mechanisms for parents. But the cat-and-mouse game continues.
Narayan told the Guardian that Britain's advantage lay in watching Australia stumble first. "One benefit of going second is that we learn some practical implementation lessons," he said. The key question, he noted, was determining which platforms to restrict and how to verify age in a way that actually works. He spoke of "robust age assurance" as an ongoing challenge, not a problem to be solved once and forgotten. The Australian experience had shown that this was not a one-time fix but a continuous process of policy adjustment and regulatory pressure.
What surprised Narayan, though, was something less tangible than compliance rates. He noted that Australia's ban had shifted the national conversation itself. Parents, families, youth workers, and carers were suddenly talking about teenage social media use in ways they hadn't before. The law had become a cultural moment, not just a regulatory one. "I really appreciated the way in which Australia pioneering in this regard has meant that parents, families, carers, youth workers are all engaging in this national conversation," he said. That shift in public awareness, he suggested, might matter as much as the enforcement mechanisms.
Beyond social media, Narayan and Australian officials also signed a memorandum of understanding on artificial intelligence security. Both nations agreed to share intelligence about frontier AI models and how to prevent cyberattacks that might exploit them. The UK has moved faster in building dedicated AI security capacity, but Narayan framed the arrangement as genuinely two-way. When asked whether tech companies might be seeking favorable regulatory treatment—on copyright, for instance—in exchange for early access to their systems, he was blunt: "Our big focus has been on shoring up British capabilities so that we are the most resilient we can be against risks. We can't just trust and rely on the companies." Security, he insisted, could not be built on trust alone.
As Britain prepares to follow Australia's lead on teenage social media, the lesson appears to be this: the law is the easy part. Making it stick, keeping up with workarounds, adjusting enforcement as companies adapt, and building public understanding—that is the real work. Australia has shown it can be done. Whether it can be done well remains an open question.
Notable Quotes
One benefit of going second is that we learn some practical implementation lessons.— Kanishka Narayan, UK online safety minister
Security is not going to be built on the foundations of trusting companies. It's going to be built by having internal capability in our society.— Kanishka Narayan
The Hearth Conversation Another angle on the story
So the Australian ban is already six months old. What's the actual picture on the ground?
Mixed. The law exists, but about two-thirds of under-16s are still on social media. The platforms are being investigated for non-compliance. It's not a failure exactly, but it's not the clean victory the headlines suggested either.
Why are so many kids still getting through?
Age verification is harder than it sounds. One platform was letting users try facial recognition 24 times a day. It becomes a numbers game. And teenagers are resourceful—they know workarounds, they borrow accounts, they lie about their age. The law assumes compliance; reality is messier.
So what's Britain hoping to learn by watching Australia struggle?
How to do it better, or at least differently. Narayan kept talking about "practical implementation lessons"—which platforms to target, how to make age checks actually stick, how to keep companies honest. He's not looking for a perfect solution. He's looking for what works incrementally.
Is there anything Australia did right?
Yes, actually. The fines—up to A$49.5 million per breach—seem to have gotten the platforms' attention. Once regulators started enforcing, companies introduced better age verification, made it harder to change birth dates, gave parents reporting tools. Pressure works.
But Narayan mentioned something else—a cultural shift.
That might be the real win. The ban forced a national conversation. Parents started talking about it. Schools engaged. It became a moment where society collectively asked: should teenagers be on these platforms? That conversation itself changes behavior, even if the law doesn't catch everyone.
And the AI agreement—is that connected?
Not directly, but it shows the same thinking: don't trust the companies to police themselves. Build your own capacity, share intelligence, stay ahead. Narayan was very clear about that. Security comes from capability, not faith.