The UK Is Banning Social Media for Under-16s. The Hard Part Is Making It Work

Britain has made its decision. On Monday morning, Prime Minister Keir Starmer announced a full ban on social media for children under 16, calling it “a big step for our country” and saying that social media “is making our children unhappy and unsafe.”

That makes the UK the second major country to go this far. Australia went first, switching on its own under-16 ban in December 2025, and its rollout has been rocky. That experience hangs over the British plan, and it is the main reason so many doubt this will work. After two years of argument, the real question in Britain is no longer whether to act, but whether a ban can be enforced.

What is actually banned, and what is not

Under-16s will be barred from holding accounts on the major platforms. The names mentioned so far are Facebook, Instagram, TikTok, Snapchat, YouTube and X, though ministers say the legally binding list will be set out later, in the detailed rules to come. Messaging apps are deliberately left out: the government confirmed that WhatsApp and Signal are not included.

The plan also restricts specific high-risk functions, even on platforms that are otherwise allowed. For under-16s, that means blocks on disappearing messages, livestreaming, and chats with adult strangers. Police and ministers link those three features directly to grooming and child sexual abuse. As Starmer put it, “Is there a situation in the offline world where you would just let your child pair up with a stranger, an adult that you don’t know anything about? No.”

A separate, wider measure targets older teenagers too. Under-18s will be barred from using romantic or sexual AI chatbots. This follows a run of incidents over the past year, including Grok, the AI chatbot built into Elon Musk’s platform X, that has been used to generate fake nude images.

Starmer made a related demand last week. He told Apple and Google to build controls into phones themselves, so children cannot take, share or view explicit images. They have three months to do it before legislation forces them. Adults would still be able to reach such content, but only after proving their age.

The timeline, and the machinery behind it

This will not all happen overnight. The government will bring the legislation to Parliament before Christmas, with protections expected to come into force in spring 2027.

Just as important as the timing is the method: rather than pass a fresh, standalone ban, Britain is using a power Parliament already granted in the Children’s Wellbeing and Schools Act 2026. (That Act became law on 29 April, when it received Royal Assent, the King’s formal sign-off that turns a bill into law.) The Act lets ministers set age and feature restrictions through regulations, and lets them adjust the list of platforms and features later without passing a new law each time.

This approach is meant to fix a problem that has dogged the debate for years: how do you define “social media” in law, and what happens when a new app appears? If the law simply named today’s apps, that list would be out of date almost as soon as it passed. So instead, ministers can add or remove platforms through regulations. If a new app takes off next year, they can bring it under the rules without passing a whole new law.

The price of that flexibility is scrutiny. Regulations are debated far less than a full law, yet they will carry the decisions that matter most: which platforms count, the precise age, and how that age is checked. All of it gets settled where fewer people are watching.

Why now, and who is for and against

Behind the decision is a national consultation that drew more than 116,000 responses between March and May, one of the largest in recent years. According to the government, 83% of responding parents said the risks of social media outweigh the benefits, and 90% backed a minimum age of 16. Starmer leaned on that, while being unusually candid about the cost. “I will not present it as cost-free, as if social media has brought no benefits to young people, because clearly that is wrong,” he said. “But government is always about choices.”

Back in 2021 we covered Facebook pausing its Instagram Kids app after a leaked internal report and a public backlash over the platform’s effect on teenagers.

Not everyone is convinced, and the objections fall into a few camps. Some question the politics and the timing. One online-harms lawyer, Mark Jones, asked why sweeping restrictions arrived only weeks after the consultation closed, and whether the evidence had been properly assessed “or whether the Prime Minister feels the need to do something radical to bolster his ratings.”

Others question whether a ban even works. Scotland’s children’s commissioner, Nicola Killean, said there is too little evidence that a ban makes children safer, and that it “does little to address underlying issues such as exploitative algorithms.” Earlier, 42 UK child-protection groups argued that a blanket ban treats “the symptoms, not the problem,” and Amnesty International called Australia’s version an “ineffective quick fix.” Even the US government, which filed a response to the UK consultation, said it preferred “narrowly targeted requirements” and that “most content should be accessible by default.”

The hard part is enforcement

Most of that scepticism traces back to Australia’s experience. Its online safety regulator found in March 2026 that around seven in ten parents said their child still had an account on a restricted platform.

The core difficulty is age verification, because proving someone is over 16 without collecting everyone’s ID or biometrics is genuinely hard. The methods on the table are a government-issued ID, a credit card check, or AI that estimates age from a face scan. Each one raises privacy and free-expression concerns. That trade-off is already playing out closer to home. We have covered how Instagram ended its encrypted direct messages for users in May; Meta blamed low take-up, but the timing tracked rising child-safety and legal pressure. That is why critics worry the cure creates a new problem: large databases of sensitive identity data, and age checks that adults must pass too.

What the ban removes, not just what it blocks

There are real losses too, as Starmer admitted: the same algorithmic feed that pushes a harmful video also serves STEM explainers, revision help, language practice, and the communities that isolated teenagers rely on. For many young people, connecting with people elsewhere in the world is now just part of growing up.

Children who are creators are a group the headlines forget entirely. An under-16 with an audience, a young musician or athlete building a following, would not be able to run their own account. A parent can operate one for them, but the child cannot manage their own presence or earn from it. These are the same young people powering a creator economy that many parents barely see, as we found when iShowSpeed toured Africa. How Britain treats them is still unanswered.

The science is also shakier than the confident politics suggests. The social psychologist Jonathan Haidt argues that platforms are driving a teen mental-health crisis, a case that has shaped laws worldwide, while researchers such as Candice Odgers and Andrew Przybylski find only small, mixed effects and no proven cause. The Electronic Frontier Foundation, a digital-rights group, argued in May that shaky evidence is fuelling a rush to legislate. So the picture is genuinely mixed: youth mental health is getting worse, but the evidence that social media is the main cause is thinner than the headlines suggest.

Apple, Google, and the shift to the device

This is why the phone itself is becoming the real battleground. At WWDC 2026 on 8 June (Apple’s annual developer conference), the company previewed its biggest set of child-safety tools in years, shipping with iOS 27 this autumn.

We have already been tracking what iOS 27 brings to iPhones. A “Child Account,” required under 13 and available up to 18, applies age-based protections across the whole system. “Ask to Browse” gates new websites. “Time Allowances” replaces blunt app limits with schedules. And “Communication Safety” now blurs violent and graphic content, not just nudity, directly on the device. Apple has also already rolled out UK age checks that ask users to verify with ID or a credit card, or accept automatic content filters.

The limitation is built in. An operating system can decide which apps a child opens, but it cannot fully police what happens inside an approved app, and many of these tools still depend on a parent switching them on. That is the same enforcement gap the ban itself faces, just moved down to the hardware.

Why this reaches Kenya too

Kenya has no minimum-age social media ban. What it has instead is a set of data-protection and child-safety rules. Under the Data Protection Act 2019, a company generally cannot process a child’s data without a parent or guardian’s consent. Everyone under 18 counts as a child under the Children Act 2022, and companies are expected to verify users’ ages. The Communications Authority adds more on top, with guidelines that tell internet and phone providers to switch on parental controls, default privacy settings and age checks.

So Kenya protects children by regulating consent and safety, not by banning access.

But the UK’s choice still reaches Nairobi. Age limits are set country by country, yet the systems built to enforce them usually are not. When Apple rebuilds how it checks a user’s age, or a platform re-engineers its systems to satisfy regulators in London and Canberra, that same design often becomes the default everywhere, Kenyan users included. We have already watched this happen: Meta recently took its strictest teen content limits worldwide, Kenya among them, after two child-safety court losses. The rules being written this week in Westminster may quietly shape what phones around the world can do, often before any local lawmaker has had a say.

What to actually watch

The ban is announced, but not yet in force. For now it is a promise, not a working system. The announcement sets the direction; the parts that decide whether it actually protects children are still unwritten. The regulations due before Christmas will carry the real detail, namely which platforms count, the exact age, and how that age gets checked. Before then, two earlier signals will show how serious the effort is: the September deadline for Apple and Google to deliver controls on explicit images, and Australia’s enforcement figures through 2026, since that is the system Britain is copying. The protections themselves only switch on in spring 2027.

For readers in Kenya, the British law changes nothing here on its own. What reaches you is the machinery built to enforce it, arriving through Apple, Google and Meta rather than through any Kenyan statute. That is the part worth watching, because it will be on your phone long before it is in any local law.