On 4 September 2026, the Committee of Ministers of the Council of Europe approved in Monaco a recommendation addressed to the governments of its 46 member States, placing education at the centre of Europe’s response to artificial intelligence.
Amid the growing number of regulations, frameworks and deadlines introduced over the past few months, this text deserves attention for a specific reason: it says, with the voice of 46 States, something that until now may have sounded like the position of a minority.
AI literacy is the foundation. And AI literacy does not mean knowing how to use the tools.
What does it actually say?
The recommendation defines AI not simply as a technology, but as a sociotechnical phenomenon: not only the systems themselves, but also the people, institutions, policies and practices that shape how those systems are designed and used.
This leads to its most important point: AI literacy means understanding the impact of AI on society, not simply knowing how to operate the tools.
The Council of Europe distinguishes this approach from other existing frameworks, which tend to place greater emphasis on technical skills.
The text provides States with a framework built around 22 guiding principles across four areas: basic understanding of AI; human rights and human dignity; democracy and the rule of law; and the capacity for social action and critical thinking.
It also places particular emphasis on the human dimension, with privacy, autonomy, equality and non-discrimination at the forefront.
The recommendation explicitly calls for AI training and continuing professional development for teachers, recognising educators as a central part of the response.
Why this matters more than it may seem
For months, we have been watching the case for AI literacy take shape from different directions.
The AI Act established it as an obligation under Article 4. The OECD and European Commission’s AILit Framework translated it into specific competencies for learners. UNESCO did the same for teachers and students. And now the Council of Europe frames it through the lens of human rights, democracy and the rule of law.
It is no coincidence that so many institutions are reaching the same conclusion at the same time.The harmful uses of AI are already happening. They have understood that the only defence that can work at scale is not to prohibit or monitor AI, but to develop the judgement of the people who use it. And there is something this convergence of standards and frameworks suggests, even if they do not always express it in these exact terms.
This goes beyond legal compliance.
AI literacy is the foundation of a coherent approach to digital education. It is also one of the few tools individuals have to navigate a world that becomes harder to understand as AI becomes more deeply embedded in everyday life.
Complying with the law is what an institution is required to do. Developing people who can navigate this world with judgement is what is really at stake. That is the point the recommendation makes particularly clear and one that educational institutions should not lose sight of.
Teaching someone how to operate a tool is one thing. Teaching them to understand what that tool does to society, who designs it, who benefits from it and who may be harmed by it is what makes them informed citizens. A student who learns only the first is left at the mercy of the second. And that distinction, which until recently may have sounded like a pedagogical conviction, is now a recommendation endorsed by almost every country in Europe.
What can an educational institution do with this?
The recommendation is addressed to governments, but its logic can be applied directly by any institution that does not want to wait for policy to catch up.
First, assess whether what is currently being taught about AI is limited to using applications, or whether it also develops an understanding of their broader impact.
Second, train teachers, the recommendation itself identifies this as a priority.
And third, develop an institutional position that treats AI as what the Council of Europe says it is: an issue that affects rights, autonomy and human judgement, not simply a catalogue of tools to learn.
The institutions already moving in this direction are not waiting for regulation to tell them where to go.
Sources
- Council of Europe, Committee of Ministers (4 September 2026). Recommendation on education as a response to artificial intelligence. Approved in Monaco. The recommendation gives effect to Article 20 of the Framework Convention on Artificial Intelligence and Human Rights, Democracy and the Rule of Law.
- OECD and European Commission (2026). AILit Framework.
- UNESCO (2024). AI Competency Framework for Teachers.
- Regulation (EU) 2024/1689 of the European Parliament and of the Council (Artificial Intelligence Act), Article 4, as applicable following Regulation (EU) 2026/1744.
