The proposal to ban social media access for under-16s is being presented as a child protection measure. On the surface, that sounds sensible. In practice, it is a political distraction. It allows ministers to appear decisive while avoiding the harder argument about why major platforms continue to profit from harm, manipulation and criminal exploitation. The rot sits deeper than age limits. The core issue is not simply whether children should be online. It is whether platforms can continue to operate as poorly governed environments where abuse, intimidation, radicalisation, sabotage and recruitment are allowed to flourish with limited meaningful consequence.
This is why the language of a simple ban is so misleading. It frames children as the main problem to be managed, rather than recognising that the digital environment itself has been designed and monetised in ways that reward outrage, speed, anonymity and reach. In safeguarding terms, that is already serious. In national security terms, it is worse. The same systems that expose children to grooming and exploitation also provide fertile ground for hostile state actors, organised extremists and opportunistic recruiters. A policy that focuses narrowly on age restrictions while leaving the architecture of harm intact is not a serious solution. It is a headline.
The recent BBC investigation involving Roman Lavrynovych, a 22-year-old recruited via Telegram, illustrates the point with disturbing clarity. This was not a case about a child stumbling across harmful content. It was a case which, according to the reporting, exposed how online platforms can be used as channels for sabotage and covert recruitment. Lavrynovych was reportedly drawn in through contact with an individual known as “EL”, said to be linked to Russian interests, and became involved in plans connected to arson at properties linked to the Prime Minister. The significance of that case lies not only in the criminal conduct itself, but in what it reveals about the operational use of digital platforms. These spaces are not merely failing to protect children. They are also being weaponised by those seeking to recruit, direct and exploit individuals for acts that undermine public safety and national resilience.
That matters because too much public debate still treats social media harm as if it begins and ends with bullying, image-sharing or excessive screen time. Those issues are real and should not be minimised. But they do not represent the full threat picture. Platforms are now part of the operating environment for influence operations, extremist networking, fraud, intimidation and sabotage. Recruitment often starts with low-level contact that appears trivial or transactional. It may involve paid tasks, private messages, coded invitations, or “job” opportunities in channels that look mundane to the untrained eye. The progression can be gradual: poster placement, graffiti, message amplification, harassment, courier work, or surveillance. In more serious cases, it can move towards arson, violence, or support for organised criminal and hostile activity. The age of the target may vary. The method remains consistent. The platform provides access, scale and plausible deniability.
Social media is therefore not just a communications tool. It is a contested space in which information can be manipulated, identities can be fabricated and communities can be deliberately set against one another. Hostile actors have long understood the value of using fake personas and engineered narratives to inflame division. That may involve posing as far-right voices, posing as Muslim voices, spreading sectarian hostility, amplifying grievance, or driving people towards more extreme interpretations of events. The point is not always persuasion in the traditional sense. Often, the aim is destabilisation. Engagement-driven algorithms then do the rest, rewarding the most provocative material and ensuring that emotionally charged content travels further and faster than measured, factual responses.
Against that backdrop, bans are weak policy. Children determined to access these services will often find workarounds, whether through VPNs, false dates of birth, borrowed devices or migration to less visible platforms. The likely effect is not the elimination of risk but its displacement. Some young people may be pushed away from mainstream services into more opaque, less moderated spaces where exploitation is harder to detect and support is less accessible. That is not safeguarding by design. It is risk relocation. Serious regulation must therefore target the systems, incentives and governance failures of the platforms themselves rather than pretending that the user can be legislated out of danger.
What is needed is platform accountability in a much more robust sense. If a service enables recruitment pathways for sabotage, extremist grooming, targeted abuse or organised disinformation, it should face consequences that are legal, financial and operational. That means higher expectations around moderation capacity, faster disruption of harmful networks, stronger identity and traceability measures where proportionate, and far greater transparency about how recommendation systems prioritise content. It also means ending the comfortable fiction that platforms are neutral hosts with little responsibility for the environments they have deliberately engineered. If they profit from attention, they must be accountable for the foreseeable harms produced by the systems used to capture it.
This is where Safe-AI principles become essential, particularly Accountability and Safety. Accountability means that platforms should be answerable for how automated systems amplify material, connect users, recommend communities and fail to interrupt known patterns of exploitation. Algorithms should be auditable. Decision-making processes should be open to scrutiny. Routes for challenge and redress should exist when harmful failures occur. Safety means designing systems that prioritise human wellbeing over frictionless engagement. Features that encourage compulsive use, endless escalation and impulsive sharing should not be treated as harmless design choices when they clearly intensify vulnerability and exposure. Safety must be built into the architecture, not pasted on afterwards as public relations.
The national security dimension must also be stated plainly. This is not just a safeguarding issue. It is a defence and resilience issue. Sabotage can be organised in direct messages. Extremist narratives can be spread and normalised through algorithmic recommendation. Hostile influence can be seeded through coordinated inauthentic behaviour. Financial transfers, including those involving cryptoassets, can support networks operating beneath the surface of apparently ordinary online activity. The United Kingdom should start treating major digital platforms as part of its critical social infrastructure, because that is what they have become. Their failures do not remain online. They spill directly into streets, schools, institutions and communities.
That is why empowering parents matters more than symbolic bans. Families need practical, technical tools that give them real control over the digital environment in the home. That includes device-level restrictions, router and WiFi controls, app permissions, monitoring settings, timed access and stronger literacy around the signs of online grooming, manipulation and ideological recruitment. Parents should not be left with vague warnings and political theatre. They should be given workable measures they can actually use. Alongside that, schools and organisations need support to build resilience, strengthen faith literacy, challenge hate-fuelled narratives and understand how online harms intersect with safeguarding, community cohesion and extremism prevention.
True safety will never come from simply banning the vulnerable while leaving untouched the systems that host, amplify and profit from the rot. If platforms can be used to groom children, spread extremist narratives, facilitate sabotage and recruit the vulnerable into criminal or hostile activity, then the central duty of government is not to chase headlines with blunt restrictions, but to impose serious accountability on the companies that built and benefit from those conditions. Safety requires regulation with teeth, technical safeguards that work, and a clear refusal to let platform providers evade responsibility for the harms embedded in their own design.
For further advice and support, please get in touch.
Soulful Spaces 