AI literacy has entered the European educational agenda. Alongside the obligation placed on providers and deployers by Article 4 of the AI Act, a shared framework is taking shape on how to introduce AI into school, from primary and secondary education to early childhood. This contribution gathers the main references, without claiming to be exhaustive.
What AI literacy means
AI literacy refers to the set of knowledge, skills and attitudes that allow people to understand, assess and use AI systems consciously. It does not coincide with the technical ability to code alone: it includes grasping the systems’ limits, critically evaluating their outputs and being aware of their ethical and social implications.
The European AILit Framework and its four competence domains
The framework “Empowering Learners for the Age of AI” (AILit), developed jointly by the European Commission and the Organisation for Economic Co-operation and Development (OECD) with the support of Code.org, presented for consultation in May 2025 and finalised in 2026, defines AI literacy for primary and secondary education. It organises competences into four domains (Engage with AI, Create with AI, Manage AI, Design AI), broken into 22 competences, and will feed the PISA 2029 assessment on media and AI literacy. Its design builds on earlier frameworks, in particular the European Digital Competence Framework for Citizens (DigComp) and the UNESCO frameworks.
DigComp 3.0 and the UNESCO frameworks
On citizens’ digital competence, the fifth edition of the European framework, DigComp 3.0 (Joint Research Centre, JRC, 2025), integrates AI competence transversally and, for the first time, introduces over 500 learning outcomes, reflecting the spread of generative systems and growing exposure to misinformation. Internationally, the United Nations Educational, Scientific and Cultural Organization (UNESCO) has published frameworks dedicated to AI competencies for students and for teachers, which complement the European action.
From theory to the classroom: early childhood resources
Translating these frameworks into practice runs through teaching resources. In Early Childhood Education and Care (ECEC), materials circulate that introduce basic concepts through play and experimentation, with activities inviting children to ask whether an AI system is always reliable, whether a machine can «think» like a person, or to tell apart what a human does from what an automated system does. They are examples of how literacy can start early, in age-appropriate forms.
Outlook
The picture is in motion: DigComp 3.0 has just been published, AILit was finalised in 2026 and national guidance on the use of AI in schools is multiplying. The direction, however, seems set: AI literacy is becoming a stable component of educational paths, from early childhood to secondary school, as basic digital competences already have.
In dialogue with the 2030 Agenda
Goal 4 (Quality education). AI literacy fits within the commitment to ensure inclusive and equitable quality education and learning opportunities for all, extending the competences needed to take part in a society shaped by artificial intelligence. Goal 10 (Reduced inequalities). Unequal access to digital and AI skills risks widening gaps: spreading literacy from school onwards is also a matter of equity.
Sources
“Empowering Learners for the Age of AI” (AILit), European Commission and OECD, 2026; DigComp 3.0, Joint Research Centre (JRC), 2025; UNESCO frameworks on AI competencies for students and teachers; guidance on the use of AI in schools; third-party early-childhood teaching resources, cited by way of example.
