On 15 June 2026 the British government said it would ban social media for under-16s and back that up with stronger age checks. The soothing line, tucked into the official fact sheet, is that many adults may not need to do much: some will already have old enough accounts, connected credit cards, or email addresses verified elsewhere. That sentence is the tell. It concedes the real point at once. Adults are inside the machinery too. The only question is which adults can glide through the gate without quite noticing it.
The child-safety case itself is not imaginary. Ofcom said in March that 72% of British children aged eight to twelve are still using services whose minimum age is nominally thirteen. The regulator has already demanded effective minimum-age policies, failsafe grooming protections, safer feeds, and an end to product testing on children from Facebook, Instagram, Roblox, Snapchat, TikTok, and YouTube. Earlier guidance on highly effective age assurance was built around pornography and other harmful material, but the principle has escaped the porn vestibule now. It is moving toward the front entrance of the ordinary internet.
A product failure becomes an identity problem
This is what happens when platforms refuse the more boring form of responsibility. They do not want slower feeds, stricter defaults, friction on stranger contact, or hard limits on recommendation systems because those measures cut directly into time-on-platform and advertising yield. So politics reaches for the thing that feels administratively neat: if we cannot trust the product, we will sort the user. If we cannot make the room civilised, we will build a guard at the door.
The trouble is that there is no children’s gate that does not classify adults. A system capable of establishing whether somebody is under thirteen, under sixteen, between sixteen and seventeen, or safely over eighteen is not a special exception layered over the edge cases. It is general social infrastructure. The architecture does not care whether the next use-case is pornography, gambling, AI chatbots, alcohol delivery, livestreaming, or something yet to be moralised into urgency. Once the proof rail exists, the next door needs only a new threshold.
Regulators are starting to admit as much in the fine print. In their joint March 2026 statement, Ofcom and the ICO said that even services that merely state in their terms that under-13s should not be there must deploy robust age assurance or else behave as though children are already present on the service. That is a profound shift. It means the old shrug of the web platform, the lazy little box asking users to self-declare a birthday and lie through their teeth, is no longer enough. Either you classify first, or you design for the assumption that children are already in the room.
Europe builds the elegant checkpoint
Brussels, naturally, prefers prettier language. The European Commission’s age-verification solution became feature-ready on 15 April 2026. The official description is full of the expected civilised promises: privacy-preserving, anonymous, user-friendly, interoperable with future EU Digital Identity Wallets. A user can prove they are over eighteen without handing over every other detail of their life. Good. That is plainly better than a grubby biometric dragnet run by the first vendor who bought a minister lunch.
But let us not become silly. A privacy-preserving checkpoint is still a checkpoint. More revealing still, the Commission notes that the same tool can be adapted not only for 18+ use-cases but also for 13+ or 65+. That is not incidental flexibility. It is the entire political meaning of the thing. What arrives dressed as a narrow child-protection solution is already being built as reusable credential infrastructure, ready to snap into the European wallet stack by the end of 2026. The “mini wallet” is not simply a shield for minors. It is a general admission token in search of new doors.
The app store becomes a border post
The cleanest example of spillover is not pornography at all. It is the app store. In October 2025, Apple warned developers that Texas’s new law would require the collection of sensitive personal information to download any app, even if a user simply wanted to check the weather or sports scores. Since 1 January 2026, new Apple accounts created in Texas have had to confirm whether the user is eighteen or older; under-18 accounts must join Family Sharing; parents or guardians must consent to App Store downloads and purchases; developers must integrate with new age-category tooling.
There it is in plain form. The age gate does not stay at the adult-content perimeter. It moves upstream into account creation, marketplace access, API design, and ordinary software distribution. Once child safety is expressed as a compliance duty, the most economical place to enforce it is never the messy edge where a particular harm occurs. It is the shared checkpoint further back. The border post wins because it is cheaper than governing the city.
Why this argument keeps winning
The politics are almost too obvious to mention. Children are the one universally legitimate route by which broad identity infrastructure can be normalised for everybody else. Sell adults on ambient credential checks for efficiency or public order and you will meet suspicion. Sell the same checks as the unfortunate price of preventing grooming, self-harm feeds, sexualised chatbots, and compulsive platform design from reaching children, and the room goes quiet. Nobody wants to be the man objecting to child safety while the platforms behave like crack dealers in a schoolyard.
Which is why the serious objection is not that children should simply fend for themselves. The serious objection is that credential infrastructure is becoming a substitute for moral, product, and institutional judgment. The platforms had easier reforms available to them years ago: chronological feeds by default for minors, autoplay throttled or removed, stranger messaging switched off, recommendation systems made narrower and slower, genuine liability for knowingly addictive design. Most of that is ugly to shareholders. An age-assurance industry, by contrast, allows the cost of failure to be pushed onto users, parents, app stores, and document-check vendors.
That is why this story will not end with a neat adult exception carved out for the innocent majority. The more the web refuses restraint, the more every government will look for enforceable thresholds, and the more enforceable thresholds will require reusable proofs. Different jurisdictions will do it with different manners. Britain does it with ministerial earnestness and Ofcom bulletins. Brussels does it with anonymous proofs, interoperability, and the sacrament of technical standards. Texas does it with a state law so blunt that Apple felt obliged to point out the absurdity of age-checking a weather app. The style differs. The underlying move does not.
The door we are building
I would still rather have the European version than the clumsy one. I would still rather have minimised data collection than a free-for-all of face scans and document hoovering. But that preference should not be confused with innocence. The internet is being rebuilt around proof at the door because the institutions that run it would rather classify users than cultivate character, limit compulsion, or govern themselves with a shred of restraint.
The web had a chance to become adult by governing itself. It chose compulsion, deniability, and scale. So now the bouncer is being hired, and he will not stand only at the children’s entrance.
Sources
UK Government, Fact sheet: new rules to protect children online, 15 June 2026.
gov.uk / new rules to protect children online
Ofcom, Keep underage children off your platforms, Ofcom tells tech firms, 6 March 2026, and Age checks to protect children online, 16 January 2025.
ofcom.org.uk / keep underage children off your platforms
ofcom.org.uk / age checks to protect children online
Ofcom and the Information Commissioner’s Office, Joint Statement on Age Assurance, March 2026.
ofcom.org.uk / Ofcom-ICO joint statement PDF
European Commission, The EU approach to age verification, updated 2026, and European age verification app to keep children safe online, 15 April 2026.
digital-strategy.ec.europa.eu / EU age verification
commission.europa.eu / European age verification app
Apple Developer, New requirements for apps available in Texas, 8 October 2025.
developer.apple.com / new requirements for apps available in Texas
Image: original editorial illustration created for HW, June 2026.