UK Government Pledges Social Media Restrictions for Under-16s by Year's End

The Children's Wellbeing and Schools Act received Royal Assent on 29 April 2026, granting UK ministers authority to restrict social media access for children under 16 [1]. Education Secretary Bridget Phillipson has confirmed the government will impose restrictions "regardless of the consultation's outcome" [2]. The consultation itself closed on 26 May 2026 [3]. Now the question shifts from whether to how—and whether the enforcement mechanism can succeed where Australia's has already faltered.

The Legal Architecture: Ministerial Power Without Parliamentary Vote

Part 3 of the Children's Wellbeing and Schools Act does not itself ban social media for children. Instead, it grants the Secretary of State powers to lay regulations—statutory instruments—that can restrict access to websites, social media platforms, apps, and games for under-18s [4]. The government must publish a progress report within three months of Royal Assent and has 12 months (extendable by six months in exceptional circumstances) to lay the implementing regulations [2].

This structure means the specific restrictions will not require a fresh Act of Parliament. The Open Rights Group has warned that "this broad amendment takes power away from Parliament and Ofcom and hands it to ministers," noting that current or future governments could restrict content they are ideologically opposed to [4]. The powers extend beyond social media to VPN access restrictions and digital curfews [4].

The government's consultation sought views on minimum age requirements for platforms, restrictions on "addictive features," mandatory overnight curfews, and the role of age assurance technologies [3]. But the decision to impose restrictions was taken before consultation responses were received—a sequencing that critics note undermines the process's legitimacy [2].

Who Uses What: The Scale of the Problem

Ofcom's 2025 data shows 95% of UK 13-to-15-year-olds use social media, with 96% having their own profile [5]. Children aged 13–14 spend almost four hours online daily [5].

Source: Ofcom Online Nation Report 2025

Data as of Dec 10, 2025CSV

YouTube dominates at 96% usage among 8-to-14-year-olds, followed by WhatsApp (63%), TikTok (58%), and Snapchat (55%) [5]. Children spend roughly 48 minutes daily on YouTube and 45 minutes on Snapchat—together accounting for half their total online time [5]. Any "regulated user-to-user service" could fall within scope, which includes Instagram, TikTok, Snapchat, Facebook, X, YouTube, Discord, Twitch, Reddit, and many community forums [6].

The government's own consultation document acknowledges that the UK has approximately 4.7 million 13-to-15-year-olds who would be directly affected [3]. The Centre for Social Justice found that 800,000 children under five already use social media [7], indicating that the problem extends well below the proposed age threshold.

Despite these figures, Ofcom's research found that nine in ten children aged 8-17 say they are happy with what they do online [5]—a data point conspicuously absent from ministerial statements supporting the ban.

Australia's Experiment: A Cautionary Dataset

Australia enacted the world's first legislative ban on under-16 social media use, effective 10 December 2025 [8]. Four months later, the results are sobering.

Source: Molly Rose Foundation / ITV News

Data as of Apr 25, 2026CSV

A Molly Rose Foundation survey of 1,050 Australians aged 12-15 found that 61% who had social media accounts before the ban still retained access to at least one platform [9]. Only about one in four 14-to-15-year-olds are complying [10]. Approximately two-thirds of young users reported that platforms took "no action" to remove or deactivate pre-existing accounts [9], prompting Australia's internet regulator to launch investigations into the five largest platforms.

Teens have employed creative circumvention methods: using parents' facial recognition credentials, purchasing mesh face masks to defeat biometric checks, deploying VPNs to mask their location, and simply lying about their age [9][10]. The platforms initially removed around 4.7 million accounts belonging to under-16s, but many users simply recreated profiles [8].

A peer-reviewed analysis published in the Journal of Public Health Policy in 2026 examined Australia's approach as "a world-first social media ban" and documented the gap between legislative ambition and enforcement reality in a globalised internet environment [11].

Norway and France: Parallel Tracks

Norway announced in April 2026 that it will introduce legislation banning social media for children under 16, with the bill to be introduced to parliament by the end of 2026 [12]. The Norwegian approach places responsibility on technology companies for age verification at login. Initially Norway proposed an age threshold of 15 when the bill went to public consultation in June 2025, but raised it to 16 to align with Australia [12].

France has piloted age-verification requirements, and several European countries—including Denmark, Spain, and Portugal—are considering similar models [12]. However, none of these jurisdictions has yet produced published data on compliance rates or measurable changes in youth wellbeing, making it impossible to assess effectiveness beyond Australia's example.

The Evidence Base: Correlation, Causation, and Contested Science

Academic research on social media and adolescent mental health has exploded—from 3,366 papers in 2011 to over 44,000 in 2025 [13].

Source: OpenAlex

Data as of Jan 1, 2026CSV

Yet the evidence for a causal link remains contested. A 2025-2026 meta-analysis of 24 studies found a "significant and positive association" between social media risk factors and various disorders in adolescents, but noted high heterogeneity across studies [14]. Longitudinal surveys provide "mixed or null evidence" on directionality—whether social media causes poor mental health, or whether already-struggling teenagers seek out social media [14].

One experimental study found that a one-week social media detox reduced anxiety symptoms by 16.1%, depression by 24.8%, and insomnia by 14.5% [15]—offering stronger causal evidence for short-term effects. But researchers caution that short-term withdrawal studies cannot establish whether permanent restrictions would produce sustained benefits or whether compensatory behaviours (increased gaming, isolation) would emerge.

Cross-sectional study designs—the majority of the published literature—cannot establish causation, a limitation that several researchers whose work is cited by ministers have publicly acknowledged [14]. The confounding role of economic stress, pre-existing mental health conditions, sleep deprivation, and reduced physical activity remains insufficiently controlled for in most studies.

Who Pays: Platforms, App Stores, or Parents?

The consultation document frames enforcement as primarily a platform responsibility. Ofcom called on six major platforms to report on their age assurance approaches by the end of April 2026 and concluded that "none" of the companies with a minimum age of 13 are currently enforcing them effectively [2].

The legislation grants ministers power to mandate specific age verification methods [4], which means platforms bear the direct compliance cost. Age verification firm Yoti—a London-based company whose CEO Robin Tombs describes it as "an important partner for the Online Safety Act"—announced in January 2026 that its MyFace facial age estimation technology achieved the highest possible testing standard (iBeta Level 3) [16]. Yoti's system can estimate age from a single selfie without storing identity documents.

For platforms, compliance costs scale dramatically with user base. Industry estimates suggest that retrofitting age assurance after launch is roughly three times more expensive than building it in from the start [6]. Small platforms and community forums—technically captured by the "regulated user-to-user service" definition—face proportionally heavier burdens than Meta or TikTok.

Specific penalty structures for non-compliance with the new regulations have not yet been published [2]. Under the existing Online Safety Act, Ofcom can impose fines of up to £18 million or 10% of global annual turnover, whichever is greater [17]. Whether equivalent penalties will apply to the new under-16 restrictions remains to be determined by the implementing regulations.

Apple and Google's app stores represent an alternative enforcement vector—age-gating at the download stage—but the consultation did not propose mandatory app store gatekeeping as a primary mechanism [3]. ISPs are not currently designated as enforcement bodies, though the ministerial powers theoretically extend to requiring VPN access restrictions [4].

The Identity Verification Industry

The age verification market stands to benefit enormously from mandatory compliance requirements. Over 400 security experts have called for a moratorium on age assurance deployment, citing privacy risks from biometric data collection [4]. The Open Rights Group notes that "this ever-expanding industry is unregulated" [4].

Yoti's positioning as a government partner predates the current legislation—the company has worked with the Online Safety Act's implementation since 2023 [16]. However, publicly available records do not show specific political donations from age verification firms to ministers in the 18 months preceding the Children's Wellbeing and Schools Act's passage. The Electoral Commission's donation registers and lobbying disclosure databases would be the authoritative sources for confirming or ruling out such connections; this investigation found no published reporting documenting them.

What is documented is a broader pattern of tech industry lobbying. OpenDemocracy has reported concerns that the legislative framework makes it "easier for Big Tech to lobby a few ministers rather than many more MPs"—a structural feature that the statutory instrument approach reinforces [18].

Who Loses: The Case Against the Ban

For LGBTQ+, disabled, and neurodivergent teenagers, social media often provides the only accessible community and peer support network. Stonewall UK provides resources specifically for disabled and neurodivergent LGBTQ+ young people, acknowledging that online spaces serve functions that physical communities cannot replicate for those with mobility limitations, social anxiety, or geographic isolation [19].

Research on LGBTQ+ youth with disabilities documents that social media serves a "two-fold benefit": spreading awareness and connecting individuals with similar backgrounds who cannot find community support locally [20]. For neurodivergent young people who find in-person social situations challenging, platforms like Discord and Reddit provide structured interaction that reduces the sensory and social demands of face-to-face communication [20].

The Open Rights Group raised a specific scenario: a future government could mandate ID checks to access LGBTQ+ content under the broad ministerial powers granted by the Act [4]. This is not hypothetical scaremongering—the powers as drafted contain no content-specific limitations.

A youth-led framework developed by the American Public Health Association in 2025 argues that teen mental health interventions should centre young people's own expertise about their digital lives, rather than imposing blanket restrictions that remove both harmful and beneficial content simultaneously [21].

Legal Challenges: Article 10 and Proportionality

Any social media ban must operate within the European Convention on Human Rights framework. Article 10 protects freedom of expression; Article 8 protects privacy. Legal analysts note that courts are likely to ask whether less restrictive measures—such as stricter enforcement of existing age limits or mandatory child-friendly design—could achieve similar outcomes [22].

Among lawyers specialising in online safety, there is "broad sympathy for the objective—but scepticism about whether a standalone ban would deliver the structural change Parliament appears to want" [22]. The proportionality test requires demonstrating that restrictions are necessary in a democratic society and that no less intrusive alternative exists.

The EFF, Open Rights Group, Big Brother Watch, and Index on Censorship jointly called on the UK government in December 2025 to reform or repeal the Online Safety Act, arguing that mandatory age verification creates surveillance infrastructure that affects all internet users, not just children [23]. The Children's Wellbeing and Schools Act's additional powers amplify these concerns.

Timeline and What Happens Next

The government must publish its progress report by late July 2026. Implementing regulations must be laid by April 2027 at the latest, though ministers have stated they intend to act by the end of 2026 [2]. The consultation response is expected in summer 2026, after which draft regulations would need to be prepared, scrutinised, and approved—likely through the affirmative procedure requiring a vote in both Houses.

Whether that timeline is credible depends on the complexity of the chosen mechanism. If the government opts for platform-side age verification mandates enforced by Ofcom, implementation could proceed relatively quickly using existing Online Safety Act infrastructure. If it requires new technical standards, accreditation of age verification providers, or app store cooperation, delays become more likely.

Norway's parallel timeline—legislation introduced to parliament by end of 2026—means the UK and Norway may effectively be racing to become the first European jurisdiction to implement under-16 restrictions. Australia's experience will loom over both processes, an uncomfortable precedent that demonstrates the gap between legislative intent and technical enforcement in an internet built without age barriers.