Children and AI: How Do We Protect Young People From New Threats?

Government consultations on banning social media for under-16s reflect broad concern about digital harm. A team of specialists in neuroscience, ethics and society (Neurosec) at the University of Oxford argues that such reviews should also address whether young people should be given access to generative AI.

With social media, ministers and commentators have focused on features such as addictive feeds and age restrictions. Our research group's work with young people shows that we must also weigh these considerations, among many others, in an era of AI-driven technology. Concerns about mental health, social comparison and addictive design certainly apply when we think about young people's online experience, but the digital world of 2026 includes far more than Instagram and TikTok. AI-based chatbots are increasingly present in young people's lives across a wide range of areas, from education to socializing.

Adolescence is a key stage for developing social understanding, a sense of identity and more. This raises pressing questions: at what age should young people have access to AI that imitates friendship or intimacy? What safeguards are needed to protect young minds from manipulation and dependence rooted in artificial "connection"?

The impact of AI on children and teenagers is a new and poorly studied field. Even so, it is already clear that AI can affect their development both positively and negatively. On the one hand, AI can provide access to educational resources, help develop creative skills and even offer support with personal problems. On the other hand, AI can foster dependence, manipulate young minds and create an illusion of reality, which can lead to problems with social adaptation.

It is important to understand that AI is just a tool, and its effect on children and teenagers depends on how it is used. That is why clear rules and guidelines are needed for the use of AI in education and in other areas involving children. Research is also needed to better understand how AI affects young people's development.

Protecting children and teenagers from the potential threats of generative AI is therefore a complex, multifaceted task. It requires joint effort from governments, educational institutions, researchers and parents. Only then can we ensure safe and beneficial use of AI for the younger generation.

In conclusion, it is worth remembering that technology is evolving very fast, and our approaches to protecting children in the digital world must be constantly adapted. Generative AI is just one of many new technologies that can pose a threat to young people, and we must be prepared to deal with those threats.

*Meta is designated an extremist organization and is banned in Russia.