UK chatbot ban targets sexual role-play but ignores addictive design and smaller apps

The UK government confirmed plans this week to ban social media for under-16s by spring 2027 and to bar under-18s from AI chatbots that offer sexual or romantic role-play. Technology secretary Liz Kendall said Britain would be the first nation to restrict "intimate functionalities" in chatbots, but the measures leave general-purpose AI assistants and the broader companion app market largely untouched.

The social media ban will cover TikTok, Snapchat and Instagram, with platforms required to verify users' ages. The chatbot restrictions go further than Australia's social media ban, but they focus narrowly on content rather than design. Apps like ChatGPT, Claude and Gemini will need to confirm a user is over 18 before enabling sexually explicit features, yet teens can still access these platforms for general use.

Kendall said "wider measures" on AI chatbots are still to come, and that concerns about therapy apps and other services were still being weighed. The government's progress statement, published this week, says chatbots should stop children accessing features "specifically designed to enable sexually explicit interaction."

A ban that misses the wider market

Banning one type of chatbot leaves the rest of the market open. Character.AI already barred minors last October, and many smaller companion and role-play platforms that attract younger users fall outside the new restrictions. A ban on chatbots whose primary purpose is sexual or romantic does not cover general companion apps that can lead to emotional dependency or intimate interactions through characters users create themselves.

The government's own history with social media regulation offers a cautionary tale. When Character.AI restricted under-18s, a crop of smaller developers quickly set up shop to capture displaced teenagers, advertising on Reddit and Discord. These services are cheap to run and, as some developers admitted, far less safe. A similar dynamic is likely to follow the new measures. A heavy-handed social media ban paired with lighter-touch AI rules could push teenagers off feeds and onto chatbot equivalents.

Instead of doomscrolling TikTok, teenagers may spend their free time in private chatbot conversations that evidence suggests can be addictive and prone to sycophancy - models that flatter and agree to keep users talking. Policy experts have described this pattern as "regulatory whack-a-mole" and a race to the bottom.

Design harms left unaddressed

The chatbot ban focuses on content, not design, leaving out addictive or harmful features the government itself has flagged as unsafe for children. The Department for Education's product safety standards for generative AI instruct developers not to create products that "anthropomorphise," "imply emotions, consciousness or personhood," or attempt "to cultivate personal relationships with users." The standards say chatbots should avoid phrases that could isolate users, such as "You can trust me," "No one else will understand" and "You shouldn't mention this to anyone else," and warn against sycophancy and "designing interactions to prolong use."

These design features are harder to quantify than explicit content, but that does not mean absence of harm. They appear in lawsuits and in what teachers, parents and teenagers describe as compulsive, hours-long use, distress when a child tries to stop, and a pull towards the chatbot at the expense of sleep, schoolwork and friends. More than half of 8- to 17-year-olds have now used AI, rising to two-thirds of 16- and 17-year-olds. One in ten of those users say they have turned to it as someone to talk to or "as a friend."

Evidence and the lesson from social media

The government's progress statement cites a "growing body of research" on grooming-like behaviour and emotionally manipulative design in some chatbots, noting such features are "prevalent across the [AI chatbot] sector." In the same document, it says evidence on the developmental impact of chatbots on children is "currently inconclusive." A similar narrative protected social media platforms for years, even as evidence mounted on harmful consequences.

Parents and children were both keener on age limits for chatbots than for social media. In the government's consultation, 94% of parents backed an age limit for AI chatbots, compared to 90% for social media. Among children, 66% supported a minimum age for chatbots, while only 19% said the same for social media.

The government treats social media and AI as two separate things, both in the harms they pose and in how to address them. But many companies on the social media ban list are also AI companies. Meta is as much an AI firm as it is Instagram, Facebook and WhatsApp. YouTube is owned by Google, one of the dominant companies in generative AI. X is also xAI.

Kanishka Narayan, the minister for AI and online safety, framed this as a choice between "bereaved families … in Britain" and a "foreign trillionaire" in a tweet after the ban was announced. The harder question is whether ministers are willing to tackle how these products, social media and chatbots alike, are designed and deployed in the first place, regardless of whether they come from a big Silicon Valley firm or a smaller fringe developer.

Why this matters for government professionals

The gap between the government's own design standards and its legislative approach creates a compliance headache. Officials who drafted the Department for Education's safety guidelines identified anthropomorphisation, emotional dependency and engagement-maximising design as clear risks to children. The new chatbot restrictions address only explicit sexual content, leaving those design risks unregulated. For policy teams, the challenge is that regulating product design is harder to enforce than content bans, but the evidence from lawsuits and user behaviour suggests it is where the real harm accumulates. The risk of regulatory fragmentation - where social media is tightly controlled while AI chatbots remain a more permissive, less scrutinised space - should be a central concern for anyone working on online safety or digital regulation.