The Ban That Made VPNs Mandatory
What happened
Australia became the first country to ban social media for under-16s when the law took effect in December 2025. Four months later, a survey by the Molly Rose Foundation found that 61% of affected Australian teens aged 12-15 who had accounts before the ban still have access to at least one platform. Workarounds include VPNs, parent-created accounts, and alternative platforms not covered by the law. Nearly two-thirds of Australian parents say the ban has failed. Meanwhile, the UK's Children's Wellbeing and Schools Bill is advancing through Parliament with a version of the same ban, and a new US federal bill would require age verification at the operating system level for all devices, not just apps. The US bill is framed as protecting children; civil liberties groups say it would create a national identity database for internet access.
Australia ran the experiment and got the result: bans shift social media use to less supervised channels without reducing it. The UK and US are now copying the failed model, and the US version is more invasive.
The Hidden Bet
The goal of these laws is to reduce teen social media use
The visible political goal is to reduce use. The actual political function may be to give parents and legislators a visible response to a concern they cannot solve technically. A ban that fails on its stated metric can still succeed as a political gesture. The continued push despite evidence of failure suggests the political function is the primary one.
VPN workarounds make the ban unenforceable
The US OS-level age verification bill would bypass this by requiring age verification before any app installs or internet access, not just at the platform level. If that passes, VPNs do not help because the barrier is at the device, not the service. The Australian ban failed partly because it was platform-level. The US bill is designed to go deeper.
These laws primarily affect the platforms
The enforcement burden falls on parents: they must verify their children's ages, manage exceptions, and monitor compliance. The ban does not reduce parental burden; it converts existing parental discretion into a legal obligation to police their children's device use. Parents who supported the ban because they wanted help got a compliance requirement instead.
The Real Disagreement
The real fork is between treating children's online safety as a technical enforcement problem versus a social and educational one. The ban approach assumes the harm can be blocked by legal access restriction. The alternative approach, digital literacy education and parental tools, assumes the harm can be reduced by teaching children to navigate content, which requires them to have access. Both approaches require something from parents. The enforcement approach requires parents to enforce compliance with a law that is technically easy to circumvent. The education approach requires parents to engage with their children about content, which is harder. Governments prefer the enforcement approach because it appears to do something and places the burden on platforms and parents, not on government. I lean toward the education approach because it scales and survives technical workarounds. But it does not generate headlines.
What No One Is Saying
The social media platforms are not fighting the bans hard enough for the obvious reason: the bans move underage users to the gray market of workarounds where the platforms retain the users but shed regulatory accountability. A teen using TikTok via VPN is still a TikTok user, but TikTok can say they complied with the ban.
Who Pays
Teens from lower-income families
Already occurring in Australia. Will replicate in UK and US if laws pass.
VPN subscriptions cost money. Workaround literacy requires access to information about how to circumvent bans. Teens with more resources evade the ban more easily, meaning the ban disproportionately affects the demographic that needs social connection most and has fewest alternatives.
Parents in jurisdictions with strict bans
US bill is in committee; if passed, would take effect 12 months post-enactment.
If OS-level age verification passes in the US, parents must register their children's devices and verify ages for every application install. Failure to comply is a parental violation, not a child's violation.
Privacy advocates and digital civil liberties organizations
Permanent once built; infrastructure does not get dismantled.
OS-level age verification creates a government-accessible database of device ownership and age for all US internet users. The infrastructure for surveillance is built under the cover of child protection.
Scenarios
Copy-paste failure
UK and US pass versions of Australia's ban. Evasion rates in those countries match or exceed Australia's 61%. The political narrative becomes 'enforcement needs to be stronger' rather than 'the approach is wrong.' Each failure produces a more invasive follow-on.
Signal UK Children's Wellbeing Bill passes with social media ban provisions. US OS-level bill advances out of committee.
Evidence changes the politics
Australia's 4-month data becomes central to UK and US legislative debates. Lawmakers skeptical of the approach use the evasion statistics to block or weaken the bans. Alternative legislation focusing on platform design accountability passes instead.
Signal Key committee chairpersons cite Australia failure data in hearings. Legislative language shifts from age bans to algorithmic harm liability.
Surveillance infrastructure
US OS-level age verification bill passes. A federal database of device-age linkages is created. Within two years, law enforcement requests access for non-child-protection purposes. The surveillance expansion is treated as a separate policy question from the child protection rationale.
Signal US bill passes with DOJ and DHS written into the verification infrastructure as custodians.
What Would Change This
If Australia's 12-month data showed a significant reduction in teen social media use, or if the mental health outcomes improved measurably in the banned age group, the bottom line about the ban failing would require revision. The current 4-month data is clear in one direction: evasion is widespread and the ban has not reduced use.