Prime Minister Keir Starmer will outline a plan on Monday that mirrors Australia's platform list but adds a prohibition on romantic and sexual AI companions and restricts late night scrolling, with the enforcement mechanism still being finalized.
Britain is set to become the second country in the English-speaking world to bar children under 16 from mainstream social media platforms, but Prime Minister Keir Starmer's Monday announcement will also try to do something Australia's law does not: wall off minors from romantic and sexual AI chatbots, a category that barely existed when Canberra wrote its template.
Per TechCrunch's June 14 report citing the Guardian and Financial Times, Starmer will outline a policy that would block under-16s from TikTok, YouTube, Instagram, Reddit, Facebook, X, Threads, Snapchat, Twitch, and Kick. The platform list mirrors the one Australia finalized in late 2025 — the world's first country to implement such a nationwide ban, which took effect in December 2025 — and the age cutoff is the same. British officials have studied the Australian rollout closely and have been quietly consulting Canberra on age-assurance architecture, which is why the two regimes look so similar on the surface.
The UK's distinctive contribution is the AI chatbot clause. Under the proposal, anyone under 18 would be prohibited from accessing romantic or sexual chatbots, a category that includes the companion-style products that have proliferated since Character.AI and Replika went mainstream. This is the part of the plan that has no real international precedent. Regulators in the U.S. and EU have issued guidance about AI companions, but no national government has yet written a hard age bar into primary legislation. If Starmer's speech confirms the details on Monday, Britain will be first.
Gaming apps sit in a different bucket. They would not be banned outright, but titles that include chat-with-strangers features would have to strip that functionality for younger users. The proposal also targets late-night scrolling through a mechanism the Guardian's reporting leaves unspecified, and British officials have not yet described it as a hard curfew, an opt-in default, or a usage cap built into operating systems. That ambiguity is worth flagging, because it determines whether the policy reaches parental controls that already exist or tries to create new ones at the platform or OS layer.
Enforcement is the open question. The UK already has the Online Safety Act, which gives Ofcom the power to require age checks and to fine platforms that fail. The government may try to use those existing powers rather than draft a fresh statute, but the Children's Wellbeing and Schools Act also gives ministers some existing powers, and a new legislative vehicle may be needed to add the AI chatbot prohibition. Neither the speech text nor a formal policy paper has been released, so the enforcement architecture is still a moving target.
The criticism is already on the record. Privacy advocates warn that age verification at this scale means giving platform operators, or their third-party vendors, a way to identify every British child online, with the data-breach consequences that implies, and that no age-assurance method on the market is foolproof. Mental health researchers note that the evidence linking social media use to harm in adolescents is correlational, not causal, and that a blanket ban may isolate the very teens it aims to protect. The family of Brianna Ghey — the British teenager whose 2023 murder became a focal point for online-safety campaigners — has testified that harmful content significantly exacerbated her daughter's eating disorder and self-harming behavior, and has pushed for the strongest possible restrictions. Both the case for and the case against belong in a reader's working model, because this is still a proposal with a contested evidence base and unresolved implementation questions.
Internationally, the UK is joining a wave rather than starting one. Australia's under-16 ban took effect in December 2025, and France, Spain, and several U.S. states have moved on age verification for adult content and, in some cases, for social platforms. The British twist is that London is trying to extend the framework to a product category, AI companions, that no peer government has yet regulated at this level. Whether the Monday speech becomes a template of its own, or whether the chatbot piece is the part that gets negotiated away in the legislative process that follows, is the open question worth watching.