Summary

• New UK consultation data shows strong support among participating parents and carers for a legal minimum age for social media access.
• The figures are politically powerful, but the government’s own caveats show they are not nationally representative.
• Online safety policy now turns on a difficult chain of evidence, product design, age assurance, privacy, enforcement, and proportionality.

A number can be accurate and still demand caution. The UK government’s latest online safety data shows strong support among a group of participating parents and carers for a legal minimum age for social media access, but the same data also shows why consultation figures need careful handling before they become the basis for platform rules.

On 1 June 2026, the Department for Science, Innovation and Technology published transparency data from its “Growing up in the online world” consultation, which ran from 2 March to 26 May. Among parents and carers of children aged 21 or under who answered the relevant question in the full-length consultation, 89% supported a legal requirement for social media services to have a minimum age of access. Of those who supported a minimum age, 96% agreed to some extent that the age should be at least 16.

The figures capture a genuine pressure in the public debate. After years of concern about harmful content, addictive design, bullying, grooming, pornography, recommendation systems, and children’s mental health, many parents want stronger intervention. Online safety has moved out of specialist policy circles and into schools, families, policing, health, and everyday argument about how childhood should work in a digital environment.

The caveats attached to the figures are just as important as the percentages. The consultation was open to anyone who chose to respond, so the sample was self-selecting rather than nationally representative. The 96% figure is narrower again, applying only to respondents who opted into that chapter and had already supported some form of legal minimum age. Those limitations do not invalidate the data, but they change what it can fairly be asked to prove.

Consultations are not opinion polls

Public consultations are not opinion polls. They collect views, evidence, experience, objections, and proposals from people and organisations willing to engage with a policy process. They can reveal the intensity of concern and the arguments that sit behind it. They are weaker when a headline percentage starts to stand in for the whole public mood.

Online safety policy is full of that tension. Public concern is often legitimate, but concern alone does not answer the harder design questions. A legal age threshold for social media quickly becomes a set of technical, legal, and operational choices. Platforms would need to assess age, decide what evidence to request, handle users who are wrongly blocked or wrongly admitted, prevent easy circumvention, and comply without creating disproportionate privacy or security risks.

The product problem behind the policy

Those questions already sit inside the UK’s regulatory system. The Online Safety Act places duties on services in scope, including child-safety obligations and age-assurance requirements for certain harmful content. Ofcom has the power to investigate non-compliance, issue penalties, and, in serious cases, seek disruption of services that fail to meet legal duties.

Age assurance is where the policy becomes especially difficult. A system designed to protect children may require platforms to know more about who their users are. A system designed to avoid intrusive identity checks may be easier to evade. In March 2026, the Information Commissioner’s Office and Ofcom published a joint statement on age assurance, aimed at helping services comply with both online safety and data protection law.

That joint regulatory work reflects an uncomfortable reality: safety and privacy cannot be treated as separate lanes. Poorly designed safety measures can lead to excessive data collection, insecure identity handling, exclusion, or new forms of tracking. Weak privacy protections can undermine public trust in the same systems intended to protect children. A clean trade-off is easier to describe than to build.

The evidence chain is longer than the politics

The evidence problem extends beyond age checks. Policymakers have to distinguish between platform harms and wider social harms, between correlation and causation, and between measures that are visible and measures that actually change outcomes. A minimum-age rule may be popular, but popularity does not establish how well it will work, how it will be enforced, or how children will respond. A technical control may be measurable, while the underlying harm remains difficult to assess.

Platforms should not be allowed to turn methodological caution into an excuse for inaction. The direction of regulation is already clear: services are expected to assess risk, design safer systems, and produce evidence of compliance. The more serious challenge is to ensure that compliance does not become a paperwork exercise in which platforms document risk without changing the conditions that create it.

The government’s decision to publish transparency data and methodology helps expose the evidence chain. The figures show strong support among motivated parent and carer respondents for firmer age rules. They do not show what all parents think, what children think, which technical model would be proportionate, or whether a minimum age would reduce harm more effectively than stronger default settings, recommendation-system controls, better reporting, school-level interventions, digital literacy, or targeted enforcement against unsafe services.

From public concern to enforceable systems

Children’s online safety now occupies a more mature and more awkward phase of policymaking. The case for intervention has largely entered the mainstream, but each additional rule requires decisions about identity, privacy, platform liability, technical feasibility, market burden, and enforceability. Consultation data can start that conversation and expose the force of public concern. It cannot finish the job.

The next stage of online safety policy will be judged less by the strength of its slogans than by the quality of its evidence, the precision of its interventions, and the safeguards built into the systems platforms are required to deploy. Strong feelings can justify scrutiny. They cannot replace design.