Field Notes

source

Children’s circumvention behaviours online

StatusStableUpdatedStands on1 sourceSupportsnotes

An independent research report commissioned by the UK’s Department for Science, Innovation and Technology (DSIT) and conducted by BMG Research, published on GOV.UK on 14 July 2026. The preserved copy is the rendered snapshot with its .raw.html provenance companion. The report carries the standard disclaimer that its findings do not represent government views or policy.

Methodology

BMG surveyed 2,299 children aged 11 to 17 between 18 and 29 May 2026: a nationally representative sample of 2,029 plus a boost of 270 recent VPN users, weighted to be representative of UK children. Findings are self-reported by children and parents, so circumvention behaviours may be over- or under-reported. Some questions asked about “settings or tools that change a child’s apparent location” rather than naming VPNs, so that non-users were not inadvertently taught how to circumvent. Subgroup cuts include age band, gender, children with special educational needs, and financially vulnerable households.

Key findings

VPN use is common but privacy-led. 58% of children have heard of VPNs, 26% have used one in their lifetime, and 22% used one in the last three months; lifetime use rises from 17% of 11- to 12-year-olds to 31% of 16- to 17-year-olds. Among users the top reasons are keeping activity private (30%), accessing certain sites, apps, or games (27%), and watching videos from other countries (25%). 22% of users — 7% of all children — use a VPN to reach age-restricted services. 81% of users say a VPN makes them feel safer, and parents are frequently involved: 22% of child users had parental help setting up and 43% say a parent pays.

Age checks are widely met but mostly weak ones. 78% of children have encountered an age check; the most common are a tick box (57%) and date-of-birth entry (56%), which children also rate least effective. Children rate government ID upload (86%), payment card verification (74%), and facial age estimation (73%) as the most effective checks they have met.

Bypass is dominated by no-tool methods. 53% of all children have deliberately chosen a website, app, or game to sidestep a check — 37% because it had none, 34% because the check looked easy. This is the vault’s best measured instance of Displacement: the response to a gate is mostly to change destination, not to defeat the gate in place. 39% have successfully bypassed a check at least once, and 14% more have tried. Of those who bypassed, 63% posed as someone older, most often by a false date of birth (45%); 11% used a parent’s or relative’s ID. 10% of all children have bypassed a check using location-changing settings or tools. Children learn bypass methods mostly from friends (50%). The deterrents bypassers rate highest are practical: harder checks (44%), permanent account bans (40%), and notifying parents (39%).

Harms cluster after bypass. 51% of children who bypassed a check encountered at least one type of harmful content afterwards, and explicit sexual content was the type most likely to be seen only after bypassing. Children who used a VPN and bypassed show a higher association with post-bypass harms (64%) than non-VPN bypassers (43%), an association the report says age effects cannot fully explain.

The report also records Ofcom’s finding that UK daily active VPN-app users temporarily doubled to around 1.5 million after the Online Safety Act’s child-safety duties took effect in July 2025, before falling back to 900,000 by November — with the driver (children or adults) unclear. Its VPN figures sit alongside the March 2026 Ofcom children’s survey and the Childnet VPN study (fielded November 2025, published 1 December 2025).

What the report is evidence for

This is the strongest UK dataset so far on the actual mechanics of child circumvention. It corroborates the Internet Matters Online Safety Act report’s central pattern: everyday circumvention is household-level and tool-free, and VPNs are a minor route used mainly for privacy. It also documents the substitution effect: children route around checked platforms toward unchecked or weakly checked ones. Age-verification circumvention services takes that substitution as the core policy problem. The government cited this evidence when it decided not to limit VPNs; No, it’s not age verification is contemporaneous press coverage of the same study.

Built on 1 source (1 external).

Working out connections…