Something unexpected is happening in the United Kingdom. Children are leaving social media.
Not in the way regulators hoped — driven off by age gates and parental controls. They’re walking away on their own. According to The Register, new data from Ofcom, Britain’s communications regulator, shows a measurable decline in social media usage among children even as adult engagement remains stubbornly high. The findings, drawn from Ofcom’s latest annual study of online habits across the UK, paint a picture that defies the conventional wisdom about screen-addicted youth — and raises hard questions about whether the platforms themselves are becoming less appealing to the generation that was supposed to be their permanent audience.
The numbers are striking. Ofcom’s research indicates that children’s time spent on major social media platforms has dropped notably over the past year, with younger teens in particular pulling back from services like Facebook, Instagram, and X. Some of this shift is migration — kids moving to messaging apps, gaming platforms, and short-form video services that don’t fit neatly into the traditional social media category. But some of it appears to be genuine disengagement. Less posting. Less scrolling. Less time online in social contexts altogether.
And adults? They haven’t budged.
This divergence matters because it complicates the regulatory story that has dominated UK tech policy for the past several years. The Online Safety Act, which received Royal Assent in late 2023 and has been rolling out in phases since, was built on a foundation of alarm about children’s exposure to harmful content on social platforms. Ofcom, empowered as the enforcer of that legislation, has been issuing codes of practice and compliance guidance to platforms with increasing specificity. The implicit assumption was that without intervention, children would only become more enmeshed in social media’s grip. The data now suggests the kids may have gotten there first.
That doesn’t mean the Online Safety Act is unnecessary. Far from it. The children still on these platforms face real risks — from algorithmic amplification of self-harm content to contact from predatory adults. Ofcom’s own research has repeatedly documented these dangers. But the regulator now finds itself in a peculiar position: building an enforcement apparatus for a problem that is, at least in one dimension, shrinking on its own.
The reasons behind the youth exodus are multiple and intertwined. Privacy awareness among younger users has grown considerably, driven in part by high-profile data scandals and in part by school-based digital literacy programs that the UK government has expanded in recent years. There’s also a cultural factor. Being on Facebook in 2026 carries approximately the same social cachet among 13-year-olds as wearing a tie to school voluntarily. It’s a parent thing. Instagram retains more youth appeal, but even there, the gravitational pull has weakened as TikTok and platforms like BeReal and Discord have fragmented attention.
Gaming is absorbing enormous amounts of the time that social media once claimed. Roblox, Fortnite, and Minecraft function as social spaces in their own right — places where kids interact, build, compete, and communicate without the feed-based architecture that defines traditional social media. Ofcom’s data reflects this shift, showing increased time in gaming environments even as social media metrics decline. Whether these spaces are meaningfully safer is debatable. They present their own moderation challenges, their own vectors for harassment and exploitation. But they are structurally different, and kids seem to prefer them.
Short-form video deserves its own examination. TikTok’s dominance among younger demographics is well established, but the way young people use TikTok differs from how they used earlier social platforms. Consumption is heavily passive. Watching, not posting. The creator-audience dynamic is more pronounced than on platforms where peer-to-peer sharing was the core activity. This means kids are spending time on TikTok but aren’t necessarily building the social graphs and identity performances that characterized the Facebook and Instagram era. It’s entertainment, not socializing. The distinction matters for how we think about risk.
Ofcom has acknowledged this complexity. In commentary accompanying the research, the regulator noted that children’s online lives are becoming harder to categorize using the frameworks developed even five years ago. The boundary between a social media platform, a gaming service, and a messaging app is increasingly artificial. A child might spend an hour on Discord — technically a messaging platform — engaged in what is functionally social media behavior: sharing memes, posting opinions, reacting to content. Meanwhile, their time on Instagram might consist entirely of watching Reels with no social interaction whatsoever.
This blurring creates headaches for regulation. The Online Safety Act’s definitions of regulated services were written with a relatively clear taxonomy in mind. Platforms that allow user-generated content and facilitate interaction fall under its scope. But the Act’s practical enforcement depends on Ofcom’s ability to categorize services and apply proportionate requirements. When the services themselves are shape-shifting — and when children’s usage patterns don’t match the regulatory model — the task becomes considerably more difficult.
Meanwhile, adult social media usage in the UK shows no comparable decline. Ofcom’s data indicates that adults continue to spend significant time on Facebook, Instagram, X, LinkedIn, and other platforms at rates roughly consistent with prior years. Some platforms have seen modest shifts — X’s user base in the UK has been volatile since Elon Musk’s acquisition and subsequent rebrand — but the overall picture is one of stability. Adults are habituated. Their professional networks, news consumption, family communication, and entertainment are woven into these platforms in ways that make departure costly.
The generational split raises an interesting possibility: that social media as we’ve known it may be a historically specific phenomenon, peaking with millennials and declining among those who follow. This is speculative, and plenty of smart people have been wrong about predicting platform decline before. Facebook was supposed to die a dozen deaths. But the pattern Ofcom is documenting — kids drifting away not because they’re forced to but because they want to — is different from previous cycles of platform migration. It’s not just that they’re moving to the next thing. Some appear to be opting out of the category entirely.
The mental health dimension is impossible to ignore. Years of research, advocacy, and media coverage linking social media use to anxiety, depression, and body image problems among young people have had an effect on how kids themselves think about these platforms. Surveys consistently show that teenagers are more skeptical of social media’s effects on their wellbeing than adults are. They’ve internalized the warnings. Whether this skepticism is driving behavioral change or merely reflecting it is unclear, but the correlation between rising awareness and declining usage is hard to dismiss.
Schools have played a role too. The UK’s approach to digital literacy education has expanded substantially, with Ofcom itself contributing resources and research to inform curriculum development. Teachers report that students are more conversant in concepts like algorithmic recommendation, data harvesting, and attention engineering than they were even three years ago. Knowledge doesn’t always translate into action, but in this case, it appears to be a contributing factor.
For the platforms, the implications are commercial as well as regulatory. Young users are the most valuable demographic in advertising — not because they have money now, but because brand preferences formed in adolescence tend to persist. If a generation grows up without deep attachment to Instagram or Facebook, Meta’s long-term revenue trajectory in markets like the UK faces a structural challenge. The company’s pivot toward AI and mixed reality can be read partly as a hedge against exactly this scenario.
TikTok’s position is more nuanced. It retains strong youth engagement, but primarily as a content consumption platform rather than a social network. If its regulatory troubles — including ongoing scrutiny over its Chinese ownership and data practices — result in restrictions or an outright ban in Western markets, there’s no guarantee that displaced users would migrate to a competing social platform. They might simply watch less short-form video. Or move to YouTube Shorts. Or do something else entirely.
Ofcom’s findings also carry implications for the broader debate about children’s online safety that is playing out not just in the UK but across Europe, Australia, and the United States. Australia passed legislation in late 2024 effectively banning children under 16 from social media, a move that drew both praise and criticism. The EU’s Digital Services Act imposes its own set of obligations on platforms regarding minors. In the US, the Kids Online Safety Act has been the subject of prolonged Congressional negotiation. All of these efforts assume a baseline reality in which children are heavily engaged with social media and need to be protected from its harms.
If that baseline is shifting — if children are self-selecting out of these environments — the policy calculus changes. Not necessarily toward less regulation. The children who remain on these platforms may be the most vulnerable, the ones least equipped to make informed choices about their online activity. Protecting them may require more targeted intervention, not less. But the political urgency that drove legislation — the sense of an entire generation being consumed by algorithmic feeds — may need to be recalibrated against actual usage data.
There’s a cautionary note here too. Measurement is hard. Ofcom’s research methodology, while rigorous, relies on a combination of surveys, diary studies, and platform-reported data. Children may underreport their usage. They may use platforms under parents’ accounts. They may access content through browsers rather than apps, evading some tracking methods. The decline Ofcom has documented is real, but its magnitude could be overstated if measurement gaps are significant.
And the definition of “social media” itself is doing a lot of work in these findings. If a child spends three hours on Discord talking to friends, watching streams, and sharing content in server channels, is that social media use or not? Ofcom’s categorization may exclude activity that is functionally identical to what happens on Instagram or X. The decline in traditional social media use among children could partly be a reclassification effect rather than a genuine reduction in the behaviors that concern parents and regulators.
Still, even accounting for these caveats, the trend is meaningful. British children are engaging differently with online platforms than they did five years ago, and the direction of change is toward less involvement with the services that have dominated the public conversation about online harms. Whether this is a temporary fluctuation or the beginning of a durable shift will become clearer as Ofcom continues its longitudinal tracking.
For now, the regulator faces a dual challenge: enforcing a legislative framework designed for a world of maximal youth engagement with social media, while adapting to evidence that the world is changing beneath its feet. The Online Safety Act gives Ofcom substantial powers, but those powers are most effective when directed at clearly defined problems on clearly defined platforms. As children’s online activity becomes more diffuse — spread across gaming, messaging, video, and hybrid services that defy easy categorization — the regulator will need to be as adaptive as the kids it’s trying to protect.
The irony is thick. Governments spent years warning that social media was irresistible to children, that without forceful intervention the platforms would capture every waking hour of every young person’s attention. The platforms certainly tried. But the children, it turns out, had other ideas.
WebProNews is an iEntry Publication