Scroll to Content

FUTURE LEARNING LAB is a research group at the University of Agder in Southern Norway, combining with an online center and international network

Future Learning Lab Themes

Learning in a faster, globalized world

Education and learning, in higher education and elsewhere, is now challenged profoundly by technologies, policies, and cultural forces of a depth and scope that was difficult to foresee only a few years back. Today we ask ourselves how the coming of social media impacts on well-being. We ask ourselves how the world of 24/7 media changes our means and modes of acquiring knowledge and insight. More information is available than at any time in human history, and yet the “knowledge industry” is challenged by unprecedented monopoly of ownership. How do educators meet the challenges?

The printing press is often considered the single most transformative media innovation for learning and education, because it did not merely accelerate the distribution of knowledge: It democratized the foundations on which knowledge could be produced, contested, and preserved across generations. By making standardized, reproducible text widely available, it enabled mass literacy. The rise of the university as a knowledge institution, the scientific revolution, and the Reformation in sum constructed the epistemological infrastructure of modern society. One could argue that every innovation, from the textbook to the internet is in a sense a footnote to Gutenberg's disruption of religious monopoly over the written word.

The invention of HTML and hypertext shattered the linear, author-controlled logic of print, replacing it with a reader-driven architecture of knowledge where any concept could instantly link to any other — collapsing the distance between curiosity and information. This non-linear, networked structure fundamentally restructured how learning happens, shifting pedagogy from the transmission of fixed texts to the navigation of living, interconnected knowledge ecosystems. No prior media technology — not the printing press, not broadcast television — embedded the learner's agency so directly into the structure of the medium itself, making hypertext not merely a delivery tool but a new cognitive environment for education.

From the instructional diagrams and letter ornamentation of medieval manuscripts, the image has functioned as a pedagogical shortcut; compressing complex knowledge into more immediate form. Images allow the eye to grasp what the text labors to explain. Each technological leap in image reproduction, from engraving, photography, film, television to digital graphics shifted the epistemic authority of the visual. It moved it from rare, hand-crafted supplement to the dominant carrier of meaning in modern education, culminating in the screen-saturated learning environments of today. What the history of the image ultimately teaches us about education is that every time visual representation became cheaper, faster, and more accessible, it reorganized the power relations of knowledge: Who could show, who could see, and whose reality counted as worth depicting.

In pre-literate societiety the trained voice was the main technology of knowledge transmission. Knowledge passed on through speech and song. Through the invention of the phonograph, radio, magnetic tape, and digital formats, sound has always occupied a uniquely intimate cognitive channel, bypassing the abstraction of text to reach the learner through rhythm, tone, and the embodied presence of a human voice. In the 20th century classroom, radio broadcast and the language laboratory represented the first systematic attempts to industrialize sound as a pedagogical tool. Later, the Walkman and the MP3 player privatized the learning ear, untethering education from physical space long before anyone used the word "mobile learning." What the history of educational sound reveals is that hearing is never merely passive reception: It is participatory, emotional, and deeply tied to memory formation. This is why every technology that made the spoken word reproducible and distributable has carried genuine transformative potential for how human beings absorb, retain, and internalize knowledge.

Time is more than the container in which learning happens. Every educational system embeds assumptions about pace, sequence, and presence that reflect deeper ideological commitments about what knowledge is, who deserves access to it, and how human development unfolds. The industrial model of education essentially synchronized learning to the clock and the calendar, imposing standardized durations like the lesson, the semester, and the degree onto cognitive processes that are fundamentally asynchronous, non-linear, and quite individual. The most radical promise of digital and networked learning is not access to more content but the potential liberation from synchronized time. The promise is the possibility that the learner, not the institution, might finally control when, at what speed, and in what sequence understanding is allowed to emerge.

Space is never a neutral backdrop for learning. The architecture of the classroom, the arrangement of desks, the placement of the teacher at the front, the walled boundary of the school itself are all pedagogical statements, encoding relationships of power, attention, and authority into the built environment before a single word is spoken. Every major shift in educational philosophy has been accompanied by a reimagining of space. Socrates debated in open-air. Monks copied bibles in insulated monasteries. The idea of a classroom long suggested student desks in rows, until the open-plan schools of the 1970s questioned that order. Many educators know –– at least by intuition –– that the geometry of a learning environment shapes the geometry of thought itself. The digital turn has not abolished educational space but multiplied and fragmented it, distributing learning across bedrooms, cafés, and devices in ways that dissolve the institutional boundary between school and world. This in turn raises the question of whether genuine learning communities can be sustained without shared physical presence, or whether the body in space remains an irreducible condition of deep education.

Learning society´s digital future

Education systems and not the least higher education institutions were forged in a time where books, libraries and written texts provided the key learning pathways. In a sense they still do but digitally. The benefits are obvious: Access to world libraries is available at the touch of a button. The less obvious implications may want our attention in the sense that education´s digital future is not so much a question of human capacity to absorb insights, the consequences of not having to talk to a library or pick up a book physically.

Have you sensed a layer of impatience in your daily reading lately? Lack of focus, lack of attention to tackle the longer text? Might it be that something is going on that is worthy of our critical attention to our digital futures ?

Around 1990, the dominant mood in media and education circles was one of cautious, analog optimism. The Cold War was ending, democracy seemed to be winning, and Francis Fukuyama was already drafting what would become his famous argument about the end of history. The future felt knowable, and largely Western in its architecture. In education, the conversation centered on television and video as the great democratizing tools — the idea that satellite broadcasting and the VCR would finally bring quality teaching to remote and underserved communities. Distance learning meant mailed cassette tapes and printed study guides. The "global classroom" was a aspirational phrase, not a technical reality. Computers were present but peripheral — mostly used for word processing and basic programming, still largely confined to wealthy institutions in wealthy countries. What almost nobody anticipated — including the most forward-thinking media theorists — was the internet as a mass phenomenon. The World Wide Web did not yet exist. Tim Berners-Lee would propose it in 1989, but its implications were invisible to virtually everyone outside a small circle of computer scientists. The digital network was expected to be a tool for researchers, not a restructuring of the entire epistemic environment.

In short, 1990 was a moment of enormous historical confidence built on entirely pre-digital assumptions — which is precisely what makes it such a useful baseline for understanding the depth of the rupture that followed. So how how does the future develop, if we think of the next three decades from now?

A brave prediction, perhaps, would be this one: If the pattern holds, the most consequential developments of the next three decades will be precisely those we are currently underestimating or misframing — and the evidence suggests that the convergence of generative AI, biotechnology, and climate-driven mass migration will not merely change the content of education but dissolve the foundational assumptions we now build on concerning what a human learner is, what knowledge is for, and who has the authority to certify either. AI governance, epistemological sovereignty, and cultural sustainability will increasingly be at the forefront of global concern.

The most immediate concern in the present is the collapse of epistemic trust. A learning society depends on shared frameworks for distinguishing reliable knowledge from noise. Generative AI has industrialized the production of plausible-sounding misinformation at a scale and speed that existing educational institutions, media literacy frameworks, and regulatory architectures were simply not built to absorb. Close behind this is the deepening inequality of the attention economy. The same digital infrastructure that promises universal access to learning is simultaneously colonizing the cognitive conditions that make learning possible. Concentration, patience, sustained effort, and the tolerance for not-yet-knowing is already affecting the world we are now building. Underlying both is a crisis of purpose that no technology can resolve: In a moment when AI can pass every exam, generate every essay, and simulate every credential, the learning society urgently needs to answer a question it has long deferred: The challenge is not how to learn faster or more efficiently, but why human learning matters at all, and what it means to be educated rather than simply informed.

Synthesized knowledge definition and AI's transformative impact since 1945

Since 1945, the dominant postwar settlement on knowledge was broadly epistemological: Knowledge was understood as something produced through disciplined human inquiry — scientific method, hermeneutic interpretation, critical theory — verified by institutional gatekeepers like universities, peer-reviewed journals, and credentialed expertise, and grounded in the assumption that a knowing subject, with intentions and accountability, stood behind every knowledge claim. The digital revolution began eroding this settlement by democratizing access and accelerating circulation, but generative AI has struck at something deeper: It has decoupled knowledge production from human cognition entirely, generating outputs that are syntactically indistinguishable from expert knowledge but produced through statistical pattern-matching with no understanding, no intention, no experience, and no capacity for being wrong in the way that a responsible knower can be wrong and held accountable for it. What AI has exposed, in other words, is that the postwar framework was always quietly dependent on a set of assumptions — that knowledge has an author, that it emerges from a relationship between a mind and the world, that it carries the possibility of genuine error and genuine correction — and by producing fluent, confident, authorless text at industrial scale, it has not answered the question of what knowledge is so much as it has made that question, for the first time in decades, genuinely urgent again.

For Plato, truth was transcendent and it was singular. Truth was accessed through reason and dialectic. It existed independently of human opinion in the realm of Forms, where the philosopher's task was to ascend toward it through intellectual discipline. Aristotle on the other hand, grounded truth more pragmatically in correspondence, defining it as the alignment between a proposition and the state of affairs it describes, as well as the author who describes it. They both agreed on something foundational: That truth was discovered, not manufactured, and that the difference between knowledge and mere opinion was not stylistic but ontological. Western institutions like the university, the court, the press, the scientific journal were built precisely to operationalize this distinction, creating procedural architectures of verification, peer review, cross-examination, and editorial accountability that functioned as social technologies for separating truth-claims from noise, however imperfectly.

Generative AI strikes at both traditions simultaneously.

What is going on, at the deepest level, is the unraveling of the institutional infrastructure of trust that Western modernity spent roughly four centuries constructing. By this we mean the interlocking system of credentialed expertise, verified publication, accountable journalism, legal procedure, and democratic deliberation that provided societies with shared mechanisms for adjudicating competing claims about reality. Generative AI has not caused this unraveling — the erosion of institutional trust was already well advanced at the outset of generative AI model usage in the early 2020´s, but it has poured an accelerant on a fire that was already burning, by making it technically trivial to fabricate evidence, simulate authority, and produce at scale the aesthetic of credibility without any of its substance. Destroying trust in a claim now requires almost no resources, while rebuilding it requires enormous institutional investment, time, and goodwill that most societies are currently not in a position to mobilize. What makes this moment historically distinctive is not merely that people disagree about facts. That is as old as human society. The distinction is that growing numbers of people have lost confidence in the existence of any procedure, institution, or authority capable of settling factual disagreements legitimately. This is not a crisis of information but a crisis of the social contract itself.

Embrace the text – go all in

Exploring learning innovation designs across various subject fields, a key observation relates to needs and calls for variations in pedagogical philosophy: Teacher-student relations are changing, knowledge consumption habits are changing, reading is challenged by watching, and definitions of “book length” are changing.The idea of “Contact Education” refers to learning designs where students are confronted with real life problems and solutions debates that engage beyond the cognitive.

Problem-based learning. situated learning, design-thinking and digital storytelling: All examples of approaches to learning that appeals to intellectual curiosity but also affective and interactive ways of knowing.

In the widest sense, a text has always been any structured symbolic artifact that carries meaning across time and space. This has been so from the time of cave paintings and oral epics to modern legal codes and hyperlinked web pages. The important but often implicit assumption underlying such definitions was that a text is the trace of an intention, the externalization of a mind that had something to say and made choices, however constrained, about how to say it. Generative AI has severed this connection decisively: It produces objects that have every formal property of a text. like structure, coherence, style, argument, and even apparent voice. The difference is that such AI-generated text originates in no intention. It expresses no experience. And it is accountable to no one. This means that we are now surrounded by text-like objects that function socially and institutionally as texts while lacking the one property that every theory of textuality from Aristotle to current times assumed. What the age of AI forces us to confront is an uncomfortable possibility: We always thought of a text as being about an object and an implied relationship between that object and a human presence behind it; and now find that this presence can be convincingly simulated at industrial scale. We are left needing an entirely new vocabulary for distinguishing artifacts that carry genuine human meaning from those that merely perform its surface features with simulated fluency.

If a text no longer necessarily implies a human intention behind it, then reading (understood since antiquity as an act of interpretive encounter between two minds across time, the reader reaching toward the meaning a writer chose to embed) becomes something fundamentally different. It can no longer be understood as a dialogue across absence, as for instance Roland Barthes and Paul Ricoeur described it. It becomes nothing more than a sophisticated pattern-recognition exercise, directed at an object that was itself produced by pattern recognition. It becomes a mirror facing a mirror with no original human reflection between them. The cognitive and pedagogical consequences are profound: We taught generations of students to read by teaching them to ask what the author meant, why a word was chosen, what a structural decision reveals about intention and worldview — and all of these hermeneutic moves presuppose a human agent whose choices repay interpretive attention, which is precisely what an AI-generated text does not provide. Generative AI hollows out the entire tradition of reading as a humanistic practice. Hencem what this demands is not the abandonment of reading but its radical reinvention What is called for is a a new literacy that teaches readers not only to decode meaning but to interrogate origin, to ask not just what a text says but what kind of entity produced it, under what conditions, with what interests, and whether the fluency they are encountering is the trace of genuine human thought or a statistically optimized simulation.

If reading has been destabilized, writing faces an even more profound identity crisis. This is so because writing, more than almost any other human practice, has been understood as the technology of thought itself. Writing is the process by which inchoate experience becomes structured meaning. It is the process by which the writer discovers what they think through the discipline of finding words for it. This which means that outsourcing the production of text to a system that has never had an experience, never struggled with an idea, and never needed language to clarify its own interiority changes more than simply how writing gets done. It strikes at the developmental and existential function that writing has always served in the formation of a thinking. Writing was never really about producing texts: It was about producing writers. It was about cultivating people whose capacity for structured thought, self-examination, and communicative precision was built through the slow, resistant, often painful encounter with the blank page. When AI removes that resistance entirely, offering fluent output in seconds, it does not liberate the writer from drudgery so much as it amputates the very cognitive and moral struggle through which human judgment itself, is developed and refined. What the age of AI demands, therefore, is a radical reconceptualization of writing not as text production but as a practice of human presence. It calls for a renewed insistence that the value of writing lies not in the artifact it produces but in what happens to the mind that undertakes it. This in turn means that the most important question facing educators, institutions, and cultures right now is not whether AI can write, which it manifestly can, but whether we still believe that human beings need to write. The follow-up question is whether we have the philosophical and pedagogical courage to defend that belief.

If writing is the technology of thought and reading is the technology of interpretive encounter, then knowledge production has always been the collaborative, contested, institutionally mediated process that emerges from their interaction. What AI now threatens is not any single element of this process but the entire ecology of effortful human engagement that gives knowledge its epistemic authority, its social legitimacy, and its capacity to be genuinely revised in the light of genuine disagreement. The risk is not that AI will produce bad knowledge. The risk is that it it will produce an overwhelming abundance of knowledge-shaped content that is too cheap, too fast, and too friction-less to sustain the conditions that make knowledge production meaningful: The slow accumulation of expertise, the disciplinary community that socializes scholars into standards of evidence and argument, the reputational stakes that make a researcher careful, and the peer challenge that forces a claim to become more precise, more honest, and more accountable than its author might naturally prefer. What is ultimately at stake is whether knowledge remains a human practice, in the sense of being a form of collective inquiry in which minds are genuinely changed by evidence, arguments are won and lost on their merits, and the products of research carry the authority of hard-won human judgment. The alternative is that it becomes a computational service, abundant and friction-less and epistemically weightless, optimized for appearance rather than substance.

To some extent one can counter this seemingly disastrous development by emphasizing much more the value of group work and interaction. The group setting reintroduces precisely the friction, accountability, and social negotiation that solo AI prompting dissolves. Learners must then defend an AI-generated output to skeptical peers, justify why one formulation is better than another, and collectively push a solution past the threshold of plausible-sounding toward genuinely adequate. Learners are then doing exactly the kind of effortful, socially embedded cognitive work that knowledge production has always required. The limitation, however, is that confining AI to group settings addresses the symptom rather than the underlying philosophical question. It assumes that the problem is individual passivity rather than the deeper erosion of the belief that struggling with a problem alone, in writing, in thought, is itself a formative and irreplaceable human experience. A generation that only ever engages AI collaboratively may still never develop the interior intellectual resources that solitary reading, writing, and thinking have historically built. The most defensible position is probably this: Group AI engagement is a necessary but insufficient condition for a healthy learning ecology. It is necessary because it restores social accountability and collective standards of adequacy to knowledge production. But it is insufficient unless it is embedded in a broader educational culture that also fiercely protects spaces of solitary struggle, values the process of individual sense-making over the efficiency of its outputs, and teaches students explicitly why the resistance they feel when thinking hard without assistance is not an obstacle to learning but its core value.

Reclaiming ownership of knowledge in the age of AI is ultimately not a technical problem but a philosophical and political one. It requires societies, institutions, and individuals to make an explicit, countercultural commitment to the proposition that knowledge is not a product but a practice, not something retrieved or generated but something earned through the kind of sustained, accountable, identity-forming human engagement that cannot be delegated without loss. The institutional dimension is critical: Universities, schools, journals, and democratic publics need to rebuild the credentialing and verification architectures that assign epistemic authority not to outputs but to processes. They need to ask not what was produced but how, by whom, through what disciplinary struggle, in what community of critical peers, with what transparency about method and limitation, effectively reasserting the humanistic principle that the knower and the known are inseparable. This being said, the deepest reclamation is personal and existential: It is the individual decision to resist the frictionless, to choose the harder path of genuine inquiry over the efficient path of prompted fluency, to insist that one's own hard-won, imperfect, slowly-developed understanding of something is worthy.

There is also a structural argument to be made: Without a structural perspective the call for individual epistemic resistance risks becoming precisely the kind of liberal self-help narrative that leaves the actually existing power relations entirely intact. The reality today is that the five or six corporations that now control the dominant AI platforms, the algorithmic distribution of information, and the infrastructures of digital attention represent a concentration of epistemic power that makes the media moguls of the 20th century look almost quaint. We have a situation in which the tools of knowledge production, the channels of its distribution, and the architectures of human attention are simultaneously owned by the same entities, which is not a market outcome but a structural condition of cognitive capitalism that requires political economy analysis, not merely media literacy education.

Hans Magnus Enzensberger's concept of the consciousness industry was developed in 1962 but it is still uncannily prescient. He identified the systemic logic decades ago: The industrialization of mind is not incidental to capitalism but central to it. That is what is produced and distributed through media systems, and it only reflects the free play of ideas to an extent. What it also reflects is the organized reproduction of the conditions under which certain kinds of knowledge, certain kinds of subjects, and certain kinds of social relations remain dominant.

Reclaiming ownership of knowledge therefore requires not just personal resolve or institutional reform but a genuinely political struggle over the public ownership of epistemic infrastructure, including the regulation of platform monopolies, the funding of independent knowledge institutions, the legal protection of communicative sovereignty as a democratic right.

Education, technology & AI

Technologies change, but education has been driven by technology since the invention of drawing, reading and writing. From the works of Walter Ong, Marshal McLuhan and others of that time, to more contemporary authors reflecting on accelerating technological change and media convergence, the education challenge is the same: To maintain a steadfast focus on the human element and values at stake.

We are living through a period of unprecedented social change. Much of it is driven by a fast accelerating technological evolution, where artificial generative intelligence post quite existential threats to established understandings of what knowledge is. Political relations are changing unpredictably fast, as well. Accordingly, our theories, concepts and models of education are also in need of critical updates.

LAYERS OF INTERDISCIPLINARITY: THE FUTURE LEARNING LAB VISION

From Walter Ong to McLuhan, the serious thinkers on technology and education keep arriving at the same point. What is at stake in every technological shift is not efficiency or access. It is what happens to the human beings living through it. Creativity has never simply been about producing something new. It has been about producing meaning. And meaning requires a self that struggles, chooses, and takes responsibility for what it makes.

The current explosion of generative AI makes one thing uncomfortably clear. Fluency and creativity are not the same thing. A system that recombines existing patterns at breathtaking speed is doing something remarkable. But it is not creating. Not in any sense that carries human weight. Real creation is an encounter between a particular person and a medium that resists. That person has a specific life. Specific things to say. The resistance is not an obstacle. It is the point.

Our models of education need updating and the update has to start right here. Creative work is not a luxury or a special talent. It is the most basic human act of making something that bears the trace of a real presence. Much of our technological infrastructure is now organized around replacing genuine human effort with faster and cheaper alternatives. Insisting on creativity as the core of education is not nostalgia. It is a refusal to give up on something essential. Being human is worth something that speed cannot deliver. The difficulty does not diminish that worth. If anything it confirms it.

Interdisciplinarity has always promised more than it delivers. The reason is usually institutional. Disciplines protect territory. Creativity, understood the way we have been discussing it, is the natural enemy of that protection. It requires the willingness to work in uncertain ground. To not yet know. To bring one specific formation into genuine contact with another and let something new emerge from the friction. That is exactly what interdisciplinary work demands. But higher education keeps organizing itself around the efficient production of disciplinary outputs. If creativity is the core of education then interdisciplinarity is not a structural challenge. It is a creative practice. Treat it that way.

Cultural industries have always had a complicated relationship with disruptive technology. The music industry fought the MP3 and lost. Film fought television and eventually absorbed it. Publishing fought digitalization and is still fighting. The pattern is consistent. First denial. Then litigation. Then reluctant adaptation. Then a new normal that looks nothing like what anyone predicted. Education is no different. Except the stakes are higher.

The reason the stakes are higher is this: Education is not just a cultural industry. It is the industry that produces the capacity to participate in all the others. When it gets the response to disruption wrong the consequences are generational. They are also political. A learning society that outsources its critical faculties to the tools of disruption has not adapted. It has surrendered.

So how does genuine harnessing happen? It starts with clarity about what the technology cannot do. AI cannot struggle. It cannot be wrong in ways that matter. It cannot care about the outcome. These are not limitations to be engineered away. They are the precise qualities that define what human learning is for. We should build from there.

Institutions of higher education need to stop asking how AI can make existing education more efficient. That is the wrong question. The right question is what kinds of learning become more possible and more necessary in the presence of this technology? Critical judgment. Creative synthesis. Ethical reasoning. Collaborative sense-making. These are not soft skills. They are the core competencies that disruptive technology cannot replicate and therefore cannot replace.

The cultural industries that survived disruption did so by finding what was irreplaceable about the human element and building their future around it. Education must do the same. Not defensively. Not nostalgically. The technology will keep accelerating. The human element will not become less important. It will become more important. That is the opportunity. And we should work hard to take it.

The reason the stakes are higher is this. Education is not just a cultural industry. It is the industry that produces the capacity to participate in all the others. When it gets the response to disruption wrong the consequences are generational. They are also political. A learning society that outsources its critical faculties to the tools of disruption has not adapted. It has surrendered.

So how does genuine harnessing happen. It starts with clarity about what the technology cannot do. AI cannot struggle. It cannot be wrong in ways that matter. It cannot care about the outcome. These are not limitations to be engineered away. They are the precise qualities that define what human learning is for. Build from there.

Institutions need to stop asking how AI can make existing education more efficient. That is the wrong question. The right question is what kinds of learning become more possible and more necessary in the presence of this technology. Critical judgment. Creative synthesis. Ethical reasoning. Collaborative sense-making. These are not soft skills. They are the core competencies that disruptive technology cannot replicate and therefore cannot replace.

The cultural industries that survived disruption did so by finding what was irreplaceable about the human element and building their future around it. Education must do the same. Not defensively. Not nostalgically. But with genuine intellectual courage. The technology will keep accelerating. The human element will not become less important. It will become more important. That is the opportunity. Take it.

Law has always lagged technology. It´s not a flaw. It´s a feature. Law moves slowly. It is built on precedent and on the careful accumulation of tested principle. Speed is not its virtue. Stability is. But the current pace of technological disruption is not merely outrunning law. It is undermining the foundational assumptions on which law rests.

Consider authorship: Copyright law assumes a human creator. AI dissolves that assumption entirely. Consider liability: Law assumes an identifiable agent whose decisions caused harm. Algorithmic systems distribute causation across millions of micro-decisions in ways that make traditional liability frameworks almost meaningless. Consider evidence: Legal procedure assumes that documents have traceable origins and that testimony reflects genuine human knowledge. Generative AI can fabricate both with considerable plausibility. These are not edge cases. They are central to how law functions.

The deeper problem is jurisdictional. The major AI platforms operate globally. Law operates nationally. The gap between these two scales is not a technical problem to be solved by better international coordination. It is a structural crisis of sovereignty.

A handful of corporations now make decisions about the epistemic environment of entire societies with no meaningful democratic accountability to those societies.

This is where law and education share common ground. Both depend on the same foundational commitment. That human beings are accountable agents. That claims have origins. That decisions have authors. That someone can be held responsible. AI does not eliminate accountability. But it makes it very easy to hide.

The legal challenge is therefore also a political one. Regulation requires political will. Political will requires an informed public. An informed public requires exactly the kind of critical education we have been describing. Law cannot solve this alone. The response must be as interconnected as the problem.

Democracy runs ideally on informed consent. Informed consent requires a shared epistemic foundation. That foundation, while always far from perfect, is cracking. It was already under pressure, long before the coming of AI: 1) Centuries of inequality, 2) post-WW2 declining trust in international institutions., and 3) the algorithmic amplification of domestic outrage in an increasing number of countries.Just few years back. generative AI arrived and made the production of political misinformation essentially free, fast, scalable and personalized. The conditions for democratic deliberation were rendered even more strained. The same concentration of platform ownership that shapes what people read and watch now shapes what people believe is real.

Political institutions are responding. Slowly. Badly. Mostly by asking the platforms to regulate themselves. That is not regulation. That is surrender dressed up as politics, and nothing more. What is needed is simpler to state than to achieve: Public ownership or genuine public accountability of epistemic infrastructure.

And underneath all of this is the education question again. Citizens who cannot distinguish reliable knowledge from manufactured noise cannot govern themselves. That is not an opinion. It is a structural fact about how democracy works. Or stops working.

Politics without epistemology is just theater. And right now we are getting very good at the theater. It did not begin with Generative AI. The symbiosis of politics and media showed signs of failure long before. Don´t blame technology: Take a closer look at how it is used and regulated instead.

Pedagogy is not a discipline. It is that, but more. That is the first thing to understand. Pedagogy is a practice that draws on philosophy, psychology, sociology, political economy, media theory, neuroscience, ethics, concrete substance fields, and more. It always has. The problem is that most universities keep treating it as if pedagogy were a department. forgetting the other parts.

The true interdisciplinary challenge is not organizational. It is not about getting different faculties to collaborate on a curriculum. That is the bureaucratic version of the problem. The real challenge is epistemological. It is about whether educators from different traditions can genuinely learn from each other. Not just cite each other. Learn.

That requires willingness to work in uncertain ground. It requires tolerance for not yet knowing. It requires the creative courage to bring one formation into genuine contact with another and let something new emerge. Most academics are not trained for this. They are trained for depth in one direction. Interdisciplinarity demands a different kind of intellectual character.

Then there is the student. The true pedagogical challenge is not delivering interdisciplinary content. It is forming interdisciplinary thinkers. People who can hold complexity without resolving it prematurely. Who can reason across frameworks without losing critical grip on any of them. Who know that the most important problems do not respect disciplinary boundaries because reality does not respect them either.

That is what education is finally for. Not the transfer of content. Not the production of credentials. The formation of minds capable of living responsibly inside genuinely complex situations. Everything else is preparation for that. Or it should be.

21st CENTURY LEARNING