Computing education and digital skills policy

• Computing and digital skills encompass both the teaching of computing as a rigorous academic discipline (covering programming, computational thinking, and the principles underpinning digital systems) and broader digital and AI literacy that all young people need to participate confidently and critically in an increasingly technology-shaped world.
• Computing education concerns the curriculum, qualifications, and teaching of computing and computational thinking across schools and colleges.
• Digital and AI literacy - the capacity to understand, use, and critically interrogate AI - are emerging dimensions of digital competence that cut across the whole curriculum. Both are essential if all citizens are to engage responsibly with a world increasingly entangled with digital technology.
• Our recommendations call for curriculum reform, a review of qualifications and assessment options, and sustained investment in infrastructure and teacher development.

A rapid review of AI literacy frameworks

Young people across the UK are growing up surrounded by Artificial Intelligence (AI), but without the knowledge and skills to understand, question and use it responsibly. A rapid review of AI literacy frameworks, commissioned by the Royal Society in October 2025 to inform wider debate on the role of AI in the education landscape, found that while teachers and pupils are already using AI in classrooms, no agreed national approach currently exists for ensuring all young people are equipped to use AI effectively and appropriately.

The review highlights that most existing AI education efforts focus heavily on technical skills (such as coding or understanding algorithms) with less emphasis on broader societal and environmental dimensions. The UK therefore risks raising a generation of competent users of AI tools, who are unaware of the societal and other challenges of this technology that is shaping their lives. The Society continues to gather evidence to build understanding of AI literacy in education and inform future policy development across the sector.

Key findings:

• Teachers are already using AI, but 'flying blind' - Over half of UK teachers already use AI for planning and marking, yet most lack training or confidence to teach about AI itself. Teachers are improvising while policy guidance lags behind.
• Fragmented approaches across the UK - England, Scotland, Wales, and Northern Ireland are pursuing different strategies, but none provides a coherent, system-wide framework. Risk of inequity - without a shared baseline, AI literacy risks becoming dependent on local initiatives or individual teacher expertise, widening educational inequalities.

System upgrade required: creating opportunities in computing education

Despite significant changes to computing education since 2014, participation remains stubbornly low, particularly among under-represented groups. System upgrade required, the Royal Society's latest report on computing education (published in October 2025), highlights significant barriers to participation and proposes changes needed to support a skilled and inclusive workforce. The report recommends updates to the curriculum, a review of qualification and assessment options and investment into infrastructure that removes barriers to access.

The skills shortage in this area costs the country an estimated £63bn a year and prevents us from fostering the next generation of innovators. For the UK to keep up with rapid digital developments and fulfil the government’s ambition to be an 'AI superpower', students must be equipped to take advantage of the potential that technological progress offers. This means that all young people must gain digital skills from their education, though not all will go on to study computer science at an advanced level.

Not enough young people are taking up computing subjects to meet demand from employers, and the computing workforce does not yet reflect the diversity of society. By 2023, roughly one in three secondary schools still did not offer computer science as an option at GCSE; at A level, approximately two in five students were not offered the chance to study the subject.

Previous work in computing education and digital skills policy

The Royal Society has a long-standing commitment to computing education. Shut down or restart? (2012) assessed the state of computing in UK schools, making the case for fundamental reform that led to a new computing curriculum in England in 2014. The 2017 report After the reboot: computing education in UK schools examined the implementation of those reforms, finding that significant challenges in teacher supply, diversity of participation, and patchy provision remained. These reports established the evidence base upon which the Society's current programme of work builds.

The Society began convening expertise and developing its evidence base on AI in education in 2023. The roundtable ‘Education in the age of AI: developing AI-literate citizens’ (January 2025) explored what it means for young people to be literate in technologies that utilise AI and the challenges of embedding these skills in UK education.

The Society considers AI in education in three broad categories:

• Teaching about AI – Supporting young people in detecting, understanding, and critically interpreting content created by AI and how it works (AI literacy), so that they can use digital technology and media in safe, responsible, and ethical ways. These skills could be defined as technical understanding, data literacy, ethical awareness, critical thinking, and continuous learning.
• Teaching for AI – Equipping young people with the mathematical, data, and computing skills required to manage and build AI systems in the future. This includes the development of computing education, digital skills in the wider curriculum, qualifications, and careers information.
• Teaching with AI – How teachers can use AI to reduce administrative burden and create increasingly personalised learning and assessment experiences for students. This category has considerable ethical implications, including mitigating harm to students, closing or exacerbating existing inequalities, reducing plagiarism and other forms of cheating, and the skills and role of teachers.

This work is partially supported through funding from the Department for Science, Innovation and Technology.