Monster or Messiah?

Liam Wegimont

Executive Director @ Global Education Network Europe

Liam Wegimont is the Executive Director of GENE, the European network of ministries and agencies of Foreign Affairs, Development Co-operation and Education, devoted to Global Education. Co-founder of ANGEL – the Academic Network on Global Education Learning. He has been a teacher, youth worker, school principal – of Mount Temple Comprehensive School in Dublin – researcher, speaker and international public servant. In 2022 he co-led the process for a new Declaration on Global Education in Europe to 2050 The Dublin Declaration.

AI & EDUCATION — CHAPTER 3

“Pure hospitality does not ask for passports… it is not concerned with border controls, but demands…exposure to alterity; welcoming the stranger “without why.” If there is a knock on the door, you open it, without asking in advance if it is a messiah or a monster. The stranger is always strange… no stranger is too strange to be included…”
  –  Richard Kearney, Linguistic Hospitality, the Risk of Translation (2019)

AI in Education is a little like this. We are welcoming a stranger into our midst—and like welcoming a stranger into our home, we are doing so with fear and trepidation, unsure whether AI will be a savior or something altogether more troubling. Right now, we should be skeptical of those who assert either with confidence. Both Punya Mishra of Arizona State University and Andrej Karpathy of OpenAI seem to clearly agree. Mishra’s brilliant keynote at Education International’s first conference on AI in Education in December 2025, labels AI in Education as an unregulated global experiment, where we are ceding power and knowledge to a “smart, drunk, supremely confident, biased, sycophantic intern.” Karpathy warns us to “keep a very tight leash on this over-eager junior intern savant with encyclopedic knowledge but who also bullshits you all the time, has an overabundance of courage and little taste for what is good.”

AI will not be our savior, neither in general nor in Education. Those who wish us to believe so are generally selling something, and what they are selling needs to be treated with caution: politically, ethically, and educationally.

And yet. In late February 2026, I read of the Polish doctor who recently took part in Anthropic’s Hackathon and, it seems, beat 13,000 competitors. The problem he was addressing was clear and real: 80% of patients leave their doctors’ visits unclear about or having forgotten what to do next. Michał Nedoszytko MD, PhD built postvisit.ai to assist with follow-up: not AI instead of doctors, but AI addressing a genuine problem in a way that could lead to better patient care and outcomes. AI can do similar good in education: enhancing student learning, promoting collaboration, providing administrative support to teachers and other educators.

But the fundamental questions that AI raises for education are not primarily technical ones. They are philosophical. And for those, we need to go back to something more basic: What is education actually for? 

That is the argument of this piece. Navigating AI in education requires us to be clear about the purpose of education itself—and there are movements, policies, and communities of practice within education that have been grappling with precisely these deeper questions for decades. They may be our most useful guides.

The response to AI from the Education community

The response from educators has ranged across a wide spectrum. At one end are those who fear that AI will replace teachers, narrow learning, undermine the integrity of assessment, and ultimately make genuine learning impossible. At the other end are those who promote AI as the latest form of educational enlightenment: a force that is here to stay, that will address structural problems like teacher shortages, deliver genuinely personalized learning at scale, and liberate learners more effectively than current models of curriculum delivery or teacher education ever could.

Both positions contain truth. Both positions contain significant blind spots. Rather than settling for a comfortable middle ground between them, education needs to look beyond the hype and the scaremongering and ask more fundamental questions: about the nature of education, the nature of intelligence, the relationship between education and human agency, and the kind of future we are trying to build.

Education needs to ask more fundamental questions: about the nature of education, the nature of intelligence, the relationship between education and human agency, and the kind of future we are trying to build

International policy frameworks for AI and Education

Several international organizations have begun to develop frameworks and guardrails for moderating or regulating the use of AI in Education. Each has genuine strengths.

• The Council of Europe, guided by its mandate on human rights, democracy, and the rule of law, held a 2025 Conference on AI and Education. Its recommendations emphasized human accountability in all AI-in-education processes; the importance of evidence-based research into the effects of AI on learners and systems; a holistic approach that pairs regulation of AI with education about AI; urgent development of sector-specific legal instruments addressing data privacy, bias, and children’s rights; and the prioritization of children’s safety and equitable educational opportunity.
• UNESCO, the UN body responsible for Education and Culture, has developed guidance for policymakers on AI and education, exploring both its challenges and possibilities, and has published AI competency frameworks for both students and teachers.
• OECD, meanwhile, recently launched its Digital Education Outlook 2026, which argues that “generative AI has the potential to transform the quality and effectiveness of learning, as well as the productivity of education systems, provided its associated risks are carefully managed.” The research it cites suggests potential benefits including boosting engagement, supporting collaboration and enabling more personalized learning. But the OECD is also clear about the challenges: digital distraction, cognitive overload, equity gaps, and the need for robust ethical safeguards. And more besides. AI also has the power to diminish deep thought, to undermine, truncate or short-circuit learning, and to supplant necessary human capacities with the promise of superhuman information processing. The word “provided” in that OECD formulation is doing a great deal of work.

What is education for? Four models

Before we can navigate AI in education, we need to be clear about the answer to that question. And the answer is not as settled as many assume.

Let me take you on a short journey. Bring to mind, if you will, your own school days—a particular school, a particular classroom full of fifteen-year-olds, one of whom was you. It’s a wet Thursday afternoon in November. Who is the teacher? Who are you sitting beside? What subject is being taught?

Now: what did the people who ran that school believe education to be for? Not what they said they believed—what their practice revealed. Because every school operates from some operative model of education, some practical answer to the question: What is this all for?

There are four predominant models that seek to answer that question. Imagine them as the four corners of a blank page.

In the top left is Model 1: Academic Disciplines. The purpose of education is to pass on the traditions of learning, to ensure students excel in subject-based knowledge and make genuine intellectual progress. Schools working from this model focus on academic results and deep subject knowledge.

In the top right is Model 2: Personal Development. The purpose of education is to ensure personal growth, creativity, and human flourishing. Schools working from this model strive to ensure that those leaving are well-rounded human beings with the skills to thrive as individuals.

In the bottom left is Model 3: Good Citizenship and Social Integration. The purpose of education is to equip students to become decent citizens who understand how the world works and can contribute meaningfully to society. A narrower version of this model, favored in some economic policy circles, focuses more instrumentally on providing for labor market needs.

In the bottom right is Model 4: Social Justice and Social Transformation. According to this model, while the world is full of beauty, there is something deeply wrong with the way humans have ordered it. Injustice abounds. Inequality persists. Our ecosystems are being destroyed. Education, in this model, does not merely help students understand the world, but it equips them to change it, toward greater justice, more human rights for all, and greater sustainability.

These are ideal types—caricatures, even. No school on the planet resides purely in any one corner. And aspects of all four are genuinely necessary. We need people with deep subject knowledge (Model 1). We need people who flourish as human beings and form decent relationships (Model 2). We need people who can navigate the world of work and contribute to society (Model 3).

The problem with the three dominant models

I have run workshops inviting educators to place their schools on this map across more than one hundred settings—in Ireland, across Europe, and globally, in schools, youth organizations, adult education, and teacher education contexts. The finding is remarkably consistent. While schools have generally moved, over the past two decades, from something close to Model 1 toward varying combinations of Models 2 and 3, Model 4 remains, almost universally, the most neglected.

Here is the difficulty. Each of Models 1, 2, and 3—however necessary and valuable—tends, in its own way, to contribute to maintaining the world as it is. Academic excellence without a critical perspective reinforces existing hierarchies. Personal development without social consciousness can produce fulfilled individuals who never question the systems around them. Social integration without transformation produces competent participants in unjust systems.

Academic excellence without a critical perspective reinforces existing hierarchies. Personal development without social consciousness can produce fulfilled individuals who never question the systems around them. Social integration without transformation produces competent participants in unjust systems

What the world needs now—what ecology and justice and peace all demand—is education that also, and urgently, prioritizes the fourth model: education as a crucible for learning how to change the world toward greater justice, deeper solidarity, and genuine sustainability.

This is not merely an argument about educational philosophy. It is an argument about survival. And it has direct bearing on how we respond to AI.

Global Education as a resource for navigating AI

The questions now being raised about AI in education are, in many respects, the same fundamental questions that progressive educators have been grappling with for decades. They include:

• The question of ownership: Who controls AI, and what does cognitive liberty mean in an age of corporate AI? (Courtney C. Radsch)
• The question of knowledge: What are the gaps, biases, and silences in AI systems, including their well-documented tendency to reproduce white, Western, colonial perspectives?
• The question of morality: Is AI amoral, or actively immoral in its effects?
• The question of sustainability: What are the ecological costs of AI at scale?
• The question of agency: Will over-reliance on AI diminish human agency, individual and collective, over time?
• The question of wisdom: If we outsource our thinking to AI, where will wisdom come from?

These are not peripheral questions. They are central to any serious educational response to AI. And they are precisely the territory that global education movements have been navigating for decades with accumulated practice, policy and community.

Global Citizenship Education (GCE), Education for Sustainable Development (ESD) and Global Education (GE) share a common commitment: to open the eyes, hearts, and minds of learners to the realities of the world—including the realities of injustice, inequity, and discrimination—and to equip them with the skills and capacities to think critically, imagine differently, and act in favor of a better world. This is, precisely, Model 4 in practice.

These movements do not merely complement the emerging international frameworks for AI and Education. They provide the deeper philosophical foundation that those frameworks, for all their value, tend to presuppose rather than supply.

Two international frameworks pointing the way

Two significant policy frameworks have shaped the discourse on education for social transformation in recent years. Both are directly relevant to the challenge of AI.

The UNESCO 2023 Recommendation on Education for Peace, Human Rights and Sustainable Development—an update of the landmark 1974 Recommendation on Education for International Understanding—sets out policy guidance on how education should be oriented to bring about lasting peace and human development. It insists that education must equip all people with the knowledge, skills, values, and behaviors for effective participation in processes that advance peace, international understanding, the eradication of poverty, and tolerance. It emphasizes the importance of preventing violence and addressing racism, xenophobia, and all forms of discrimination. This framework now informs national strategy on global education across the world.

The European Declaration on Global Education to 2050—the Dublin Declaration—aligns with the UNESCO Recommendation and brings together the diverse fields of Global Education, Peace Education, Human Rights Education, Global Citizenship Education, Development Education, Education for Sustainable Development, Anti-Racist Education, and Intercultural Education into a shared policy framework with long-term strategies. Adopted in November 2022, it has already informed national policy and strategy in many European countries and was developed through a broad process that included engagement with practitioners in Africa, Asia-Pacific, and Latin America.

More recently, deepening global crises in peace, democracy, climate change, and social justice have underlined the urgency of this vision and the threat it faces from those who fear an educated, critically minded public. Policymakers, scholars, civil society actors, and educators are mobilizing with renewed energy to ensure that learners of all ages are equipped with the knowledge, skills, and agency to navigate today’s global challenges and contribute actively to creating a more just, peaceful, and sustainable world.

Global Education in practice: responding to AI

Across Europe, global education practitioners are already developing responses to the specific challenges that AI poses in education. Their work provides both models and resources for those navigating this terrain. Below are a few examples of how the question is being addressed at a national level.

• In Germany, the newly agreed Curriculum Orientation Framework on ESD for Upper Secondary Education is facilitating curriculum reform that will further integrate critical perspectives on technology, justice, and sustainability at the heart of upper-secondary schooling throughout 16 independent school systems across Germany. It is the result of cooperation between the Standing Conference of the Ministers of Education and Cultural Affairs (KMK) and the Federal Ministry for Economic Cooperation and Development (BMZ); coordinated by Engagement Global. https://www.engagement-global.de/de/orientierungsrahmen.  
• In Ireland, a long-term strategy is working to ensure that all teachers leaving initial teacher education are skilled in Global Citizenship Education, Education for Sustainable Development, and intercultural learning—capacities that are directly relevant to navigating the ethical and social dimensions of AI. https://thediceproject.ie/  This is one of several sector-wide strategies that form systemic support for the integration of GE across all learning.   
• In Malta, a mapping of existing provision is informing more effective implementation of global education frameworks, building a clearer picture of what is already working and what gaps remain for policymakers, teacher educators, school education officers, school leadership, and related sectors to implement—working toward the day when all people in Malta will have access to quality Global Education. Mapping_GE_Malta.pdf
• In Portugal, the recently finalized National Strategy on Development Education (ENED) provides a long-term framework for ensuring that education at all levels engages with questions of justice, inequality, and global interdependence. This “whole-of-government” approach involves all relevant government ministries and agencies committing to learning for global and local justice and sustainability.  ENED Portuguese Strategy 

(Further examples from other countries across Europe can be found at www.gene.eu; while examples from Latin America are available here Camoes Ibero America and ASIA-Pacific here: www.unescoapcieu.org)

If there is a knock at the door, you open it—without knowing in advance whether the visitor is a messiah or a monster. AI is at the door of education. We cannot refuse to open it

Conclusion: returning to fundamentals

We began with Richard Kearney’s image of hospitality: If there is a knock at the door, you open it—without knowing in advance whether the visitor is a messiah or a monster. AI is at the door of education. We cannot refuse to open it.

But hospitality without wisdom is pure recklessness. Opening the door intelligently requires us to know what our home is for: what we value, what we are protecting, what kind of future we are trying to build. It requires education to be clear about its own purposes, including—especially—the neglected fourth purpose: learning to change the world toward greater justice.

The international frameworks emerging around AI in Education are necessary. But they are not sufficient. What gives them depth, and what equips us to ask the right questions of AI, is a prior commitment to what education is fundamentally for.

The good news is that we are not starting from scratch. The movements, frameworks, and communities of practice that have been building global education for decades have been working on precisely these questions. They represent, perhaps, our most underused resource for navigating the challenge that AI now presents.

The knock at the door has come. We should open it well.

References and Further Reading

• Kearney, R. (2019). Linguistic Hospitality, the Risk of Translation. Études Ricoeuriennes / Ricoeur Studies, 10(1).
• UNESCO (2023). Recommendation on Education for Peace, Human Rights and Sustainable Development.
• OECD (2026). Digital Education Outlook 2026: Exploring Effective Use of Generative AI in Education. oecd.org
• Council of Europe (2025). Conference on AI and Education: Regulating the use of AI systems in education. coe.int
• European Declaration on Global Education to 2050 (Dublin Declaration), adopted November 2022.
• Kriss, S. (2026). Child’s Play: Tech’s New Generation and the End of Thinking. Harper’s Magazine, March 2026.
• OECD (2025). The Impact of Digital Technologies on Students’ Learning (Literature Review, September 2025).
• Radsch, C. C. Cognitive Liberty in an Age of Corporate AI.
• postvisit.ai  –  Michał Nedoszytko MD, PhD: postvisit.ai