The World's Strictest Ban: Should We Be Proud?

Since June, Malaysia's under-16s have been locked out of social media. I keep thinking about the day they're let back in.

Count the phones that made it inside the circle anyway.

Malaysia, it seems, may now have the strictest social media age rules in the world. I've been sitting with that phrase for a few months, and I still can't decide whether it is something to be proud of.

Let me back up a little.

I've been reading Jonathan Haidt's The Anxious Generation, and I'll admit it's a hard book to put down. Haidt argues that the shift from a play-based childhood to a phone-based one has rewired how young people grow up, and that the rise in adolescent anxiety and depression has followed (Haidt, 2024). Whether or not every graph convinces you, it's difficult to read it as a teacher, a researcher, or simply someone who cares about young people and not feel that something has gone wrong.

So when Australia's under-16 restrictions came into force last December, I was impressed. Here was a government willing to act now. I brought it into my class and asked my students to debate it. Should there be a ban? What does it actually mean to have one? I expected some disagreement. I didn't expect how loud it would be. These young adults, most only a few years past 16 themselves, argued passionately on behalf of the younger members of society who wouldn't get a say. Some called it overdue. Others called it exclusion dressed up as protection. Nobody was neutral.

Then on 1 June this year, it was our turn. The under-16 rules under Malaysia's Online Safety Act 2025 (ONSA) came into force, requiring platforms with at least eight million Malaysian users to verify age using a MyKad, passport or MyDigital ID.

Age verification required. Date of birth: whichever one gets me in.  Image: Shufti Pro, from "Malaysia Enforces Under-16 Social Media Ban With Age Check"

My first reaction was a quiet sense of pride. Our government had actually done something. Then, as I kept reading and following the news (and the pushback, of which there was plenty), the worries started creeping in. Should we take pride in that? Or should we take a step back?

I should say upfront that I'm not a policy expert. Social media and young people is simply a space I pay a lot of attention to, because so much of my research and teaching sits right in it: how young adults see themselves, how they compare, and what feeds and filters do to their sense of self (I've written before about fitspiration and lookism.) My students are the generation just on the other side of this line. The children this ban affects will be sitting in my lecture theatre in a few years, and I'd like them to arrive ready.

The worry behind the ban is real, and I share it. Parents aren't imagining the late nights, the constant comparison, the content no child should see. But I'm also painfully aware that the evidence is more contested than the headlines suggest. Researchers such as Candice Odgers have questioned whether social media is really the cause of declining adolescent mental health, rather than one among many (Odgers, 2024).

Just this week, defending its ban in the High Court, the Australian government conceded that there is no scientific consensus on social media harms for teenagers, while arguing that "credible risks" justify the ban anyway (Taylor, 2026). That stayed with me: even the people defending the ban don't claim the evidence is in. If the experts can't agree, I'm wary of any single policy being treated as the answer.

Does the ban even work?

From what I've read so far, the early signs from Australia aren't encouraging. A study published in The BMJ in June surveyed 408 Australian adolescents before their ban and again three months after. More than 85 per cent of those under 16 were still using restricted platforms, mostly through their own accounts, with others turning to fake accounts or private browsing (Barnes et al., 2026). It's a small, early study, and Malaysia may well do better at the front door, since ours requires official identity documents. But a stricter gate does little about accounts that already exist, an older sibling's login, or a parent's phone. The cultural anthropologist Mizuko Ito put it more bluntly than I would: "The fact that bans are failing should not be a cue to double down on enforcement. It is evidence of a misdiagnosis" (Ito, 2026).

As someone who studies faces (for a living, yes), I also have a narrower question I haven't yet seen answered: how accurate are these checks? Many age assurance systems estimate age from a selfie before asking for documents, and even before Australia's ban began, there were warnings that age-scan errors could let teens slip through or wrongly lock them out (Crivellaro, 2025). The US National Institute of Standards and Technology found that error rates vary widely from one algorithm to the next, that for Southeast Asian faces the result depends heavily on which algorithm is used, and that errors are generally higher for girls than for boys (NIST, 2024). Our new codes require platforms to assess their risks every year. Nothing I've read requires them to publish what they find. So, have we tested it? Perhaps. But nobody outside the platforms can tell us.

Then there are the effects nobody intended. ARTICLE 19, an international free expression organisation, has argued that Malaysia's approach risks cutting children off from learning, connection and self-expression, pushing them towards riskier corners of the internet, and shutting out those without proper documents (ARTICLE 19, 2026). There's a quieter cost too: the stricter the gate, the more identity data we ask children and families to hand over.

These are what I'd call unintended consequences. But the one that worries me most is a little different.

A ban is a deferral

Mizuko Ito has a line I keep returning to: social media "cannot be easily extracted from teens' lives, as if it were some kind of contaminant to be flushed from the body" (Ito, 2026).

A ban doesn't remove risk. It defers it.

Deferral only protects children if we use the waiting time to prepare them. Otherwise we've simply moved the risk to a later date, when it arrives all at once.

Think about how we teach young Malaysians to drive. Nobody hands a 17-year-old the keys and points them at the highway. There's an L licence, then a P licence and two probationary years, because we know competence is built gradually, with support, and with mistakes made while the stakes are still low.

A blanket ban does the opposite. It's zero exposure, then total exposure. On their sixteenth birthday, young people gain full, unsupervised access to an environment they've never practised in. And they get it at a stage when parents know less and less about their day-to-day lives (Keijsers & Poulin, 2013), and when peers matter more than at almost any other point in life (Albert et al., 2013; Steinberg & Monahan, 2007). That's the cliff edge.

And here's what struck me: this isn't quite what Haidt himself proposes. His prescription pairs delaying social media with giving children far more independence, free play and responsibility in the real world. Take away one kind of risk, and replace it with a better kind of practice. A ban delivers the first half. I'm not sure who's responsible for the second.

The world won't wait for them

There's a wrinkle that makes the cliff steeper. The online world doesn't stand still while children wait. A 13-year-old today turns 16 in 2029, and the feeds, filters, synthetic faces and AI companions waiting for them will be far more capable than anything around now. So the gap between what they'll need to navigate and what they've practised widens during the ban.

We're not pausing the problem. We're letting it grow while the people who'll face it are kept out of the room.

The UN seems to have seen this part coming. When the Committee on the Rights of the Child reviewed Malaysia earlier this year, months before the ban took effect, it flagged a lack of digital literacy and a significant digital divide among our children. It asked Malaysia to scale up digital literacy for children, teachers and families, and to develop rules on artificial intelligence that protect children while opening up opportunities for them (UN Committee on the Rights of the Child, 2026). The UN's worry, in other words, was preparation as much as access.

Where industry fits

I don't think the answer lies only with the government. Some platforms have built tools that look a lot like the graduated route I'm describing. I first learnt about TikTok's Family Pairing at an event the company invited me to: parents link their account to their teenager's and set limits together. Whatever we think of the platforms, that's supervised practice. The uncomfortable irony is that a child with no legitimate account has no account to pair. A blanket ban may reduce the use of exactly the kind of tool that builds readiness.

ECPAT International, whose entire purpose is protecting children from exploitation, has made a similar point. It doesn't support blanket bans as a primary policy tool, arguing that they risk shifting responsibility away from the companies best placed to drive change, and that children need "guidance, not exclusion" (ECPAT International, 2025). When those most focused on child safety are wary of bans, I think it's worth listening.

So...where do we go from here?

I wish I could tell you. I've gone back and forth on this for months. Some days I think about the parents who finally have some backup, and the children who'll be spared things they should never have seen. Other days I think about the fifteen-year-old cut out of the group chat, and the sixteen-year-old walking off the cliff alone. At this moment, I don't know where I clearly stand.

What I do know is that a ban doesn't remove risk. It only defers it.

So the questions that keep nagging at me are about the years in between. Who is preparing our children for what's waiting at sixteen, AI included? Is there room for something between nothing and everything, more like a P licence than a cliff? And if the checks get it wrong, would any of us ever know? (For what it's worth, the UN has already asked Malaysia to require companies to publish their children's rights impact assessments.)

A ban buys time. I just hope we use it well.

With reticence & cautious hope,

Jasmine

p.s. I tend to write about the things that keep me up at night, or that turn up uninvited in the shower. This one has been living in my head for a couple of months. I've wanted to write about it for a while but couldn't find the right way in, probably because I still haven't made up my mind. So please read this less as a position and more as me thinking out loud. I'd love to hear where you land!

---

References

Albert, D., Chein, J., & Steinberg, L. (2013). The teenage brain: Peer influences on adolescent decision making. Current Directions in Psychological Science, 22(2), 114–120.

ARTICLE 19. (2026). Malaysia: Mandatory age verification undermines privacy and free expression. https://www.article19.org/resources/malaysia-mandatory-age-verification-undermines-privacy-and-free-expression/

Barnes, C., Wolfenden, L., et al. (2026). Assessing the early effects of Australia's 'Social Media Minimum Age Act' on adolescent social media use. The BMJ, 393. https://doi.org/10.1136/bmj-2026-363695

Crivellaro, G. (2025, December 6). Social media ban for under 16s sparks concerns over age check errors, facial scanning risks. The Nightly (AAP). https://thenightly.com.au/australia/social-media-ban-age-check-fears-mount-ahead-of-under-16-lockout-c-20906874

ECPAT International. (2025, December 12). ECPAT International statement on social media bans for children. https://ecpat.org/ecpat-international-statement-on-social-media-bans-for-children/

Haidt, J. (2024). The anxious generation: How the great rewiring of childhood is causing an epidemic of mental illness. Penguin Press.

Ito, M. (2026, September 17). Teen social media bans miss the point. The MIT Press Reader. https://thereader.mitpress.mit.edu/teen-social-media-bans-miss-the-point/

Keijsers, L., & Poulin, F. (2013). Developmental changes in parent–child communication throughout adolescence. Developmental Psychology, 49(12), 2301–2308.

NIST. (2024). Face analysis technology evaluation: Age estimation and verification (NIST IR 8525). National Institute of Standards and Technology.

Odgers, C. L. (2024). The great rewiring: Is social media really behind an epidemic of teenage mental illness? Nature, 628, 29–30.

Steinberg, L., & Monahan, K. C. (2007). Age differences in resistance to peer influence. Developmental Psychology, 43(6), 1531–1543.

Taylor, J. (2026, September 29). Australian government concedes no scientific consensus on social media harms for teenagers, but 'credible risks' justify ban. The Guardian. https://www.theguardian.com/australia-news/2026/sep/30/social-media-ban-australia-high-court-mental-health-risks-teens

UN Committee on the Rights of the Child. (2026). Concluding observations on the combined second to fourth periodic reports of Malaysia (CRC/C/MYS/CO/2-4). https://docs.un.org/en/CRC/C/MYS/CO/2-4