Australia’s recent attempt to restrict under-16s from major social platforms exposes enforcement difficulties and highlights how peer influence impacts compliance, prompting calls for broader evaluations of online restrictions.
Australia’s attempt to keep under-16s off major social media platforms is already doing something less tidy than policymakers might have hoped: it is exposing just how difficult digital restrictions are to enforce once they meet teenage behaviour. In an op-ed for STAT, Camilla Iannone, research lead in the mental health evidence team at Wellcome, argues that this messiness is precisely why the wave of new rules now spreading across countries should be studied as a large-scale natural experiment, rather than judged only by whether individual teenagers can be blocked at the point of entry.
That matters because early evidence from Australia suggests the ban has not yet produced a sharp drop in use. A study published as an NBER working paper found that fewer than one in three 14- and 15-year-olds were complying four months after the country introduced its December 2025 restrictions, with many young users getting round age checks. The same research found that teenagers were far more likely to comply only if they believed enough of their peers were also doing so, underscoring a social dynamic that may weaken any ban that is not broadly observed. Australia’s online safety regulator has also said that more than 80% of children are still using at least one restricted platform.
The wider policy picture is shifting quickly. According to the STAT essay, Indonesia has followed Australia’s lead, and the European Union has gone further by restricting social media for children under 13 and personal accounts for those under 15. At the same time, Meta agreed in August to pay up to $18 billion to settle U.S. claims that Facebook and Instagram harmed children, while also promising new “safety by design” measures. Those include time limits for teenagers, quieter notifications during school hours, tighter controls on visible social signals and stronger age-assurance tools, according to reports in The Washington Post.
Yet even if such rules are enforced more tightly, the effects are unlikely to be uniform. Iannone argues that there is no single “social media effect” on all teenagers, because outcomes may differ according to age, existing mental health problems, sleep patterns, gender, income, geography and LGBTQ+ identity. Wellcome is planning to fund evaluations of the United Kingdom’s own ban, due to start in spring 2027, and the young people advising the project reportedly wanted researchers to look well beyond screen time. They asked what would replace social media in daily life: sport, friendship and better sleep, or boredom and a shift into less regulated corners of the internet.
Disclaimer: This content is for informational purposes only and is not intended to be a substitute for professional medical judgment, advice, diagnosis, or treatment.
