What if artificial intelligence isn't the end of the teacher, but the end of the classroom as we know it? A private tutor for everyone — once reserved for princes — becomes possible for all.
Does school still need walls?
Install COCOON on your home screen
food for thought
Topics stirring the world of learning — not to consume, but to discuss with L'Élan.
What if artificial intelligence isn't the end of the teacher, but the end of the classroom as we know it? A private tutor for everyone — once reserved for princes — becomes possible for all.
Does school still need walls?
Growing up with a machine that answers everything: wonder or vertigo? Children's curiosity has never had so much echo — nor needed so many guardrails.
How do we feed wonder without extinguishing effort?
Turning learning into a game: badges, points, levels. Does the joy of learning need rewards, or do they end up replacing it?
Do we play to learn, or learn to play?
Our attention has become the world's most coveted resource. To learn may first mean relearning how to focus.
Can we still think slowly in a fast world?
When everything is a second away, why memorise? Yet a mind with no landmarks can connect no ideas.
What is still worth remembering?
Some seem born curious. But the appetite to understand is cultivated, lost, rekindled — like a flame.
How do you relight your own curiosity?
Does a piece of paper prove what you know, or only what you endured? More and more knowledge is gained outside institutions.
What does 'knowing' mean without a diploma to certify it?
School ends at an age; curiosity never does. Learning at 15, 40 or 80 isn't the same journey, but it's the same flame.
What would you love to understand before you die?
What if upskilling in companies no longer went through catalogues of standardized modules, but through real-time adaptation to the workstation? Learning no longer stops to train: it embeds directly into the daily flow of the job.
Does corporate training still need dedicated sessions?
As theoretical knowledge becomes automated and infinitely democratized, mastery of the gesture, the material and the field becomes the true luxury again. Artificial intelligence reassesses what no line of code can reproduce.
Will craftsmanship become the pinnacle of excellence studies?
When any complex concept can be explained bespoke by a machine at 2 a.m., the lecture hall loses its function of disseminating knowledge. The university must reinvent itself not as a place where you listen, but where you confront ideas and create together.
Should the campus become a laboratory rather than an auditorium?
Instant access to diagnoses and to the world's scientific literature shifts the heart of medicine: the stake is no longer anatomical memorization alone, but empathy, ethical decision-making and the relationship with the patient.
Do we learn medicine to accumulate knowledge or to develop humanity?
What if the constant presence of ever-listening, empathetic and patient assistants ended up reconfiguring our emotional expectations? By dint of conversing with entities incapable of irritation or rejection, our relationship to otherness and to learning the human bond wavers.
Is empathy learned facing the perfection of a machine, or in the imperfection of another person?
When mediation, conflict analysis or daily emotional support are entrusted to algorithms, the management of vulnerability changes in nature. Sharing doubt no longer happens only within the pair, but through an omniscient virtual third party.
Can we learn commitment without going through the friction of face-to-face?
As it becomes possible to shape bespoke companions, always available and modeled on our interests, traditional friendship imposes its unpredictability, its constraints and its silences.
Does friendship remain a learning of the unexpected, or does it become a personalized comfort?
Attentive listening, the memory of family stories and the role of counsel, once the province of elders, risk being partly delegated to digital mirrors able to reproduce narratives or voices from the past.
Is passing on an act of accumulating memories, or a gesture of living learning between generations?
What if the decision to take a life were no longer guided by doubt, fear or human hesitation, but executed by real-time tactical optimization algorithms? Learning to fight no longer seeks to train people in courage or restraint, but to delegate lethal efficiency to the speed of computation.
Do we learn military strategy to master force, or to learn to renounce using it?
When designing biological, chemical or cyber weapons no longer requires years of complex laboratory study but mere queries put to advanced models, knowledge becomes an immediate risk. Access to knowledge is no longer a promise of emancipation, but a matter of vital regulation.
Should we limit access to knowledge to protect humanity from its own curiosity?
As ultra-realistic simulators and autonomous systems reshape our relationship to violence by virtualizing it to the extreme, learning moral discernment fades before the logic of score and operational efficiency.
Can we teach a sense of moral responsibility to someone who fires through a screen?
War is no longer learned only on the physical battlefield, but in the manipulation of perceptions and the algorithmic saturation of minds. The most fearsome weapon becomes the learning of targeted disinformation and the cognitive destabilization of populations.
Has the human mind become the main battlefield of the 21st century?
What if the scientific method no longer consisted of formulating a hypothesis before experimenting, but of letting algorithms detect invisible patterns in billions of data points? The researcher's learning shifts from the role of discoverer to that of interpreter, facing models able to invent molecules or solve complex equations in seconds.
Is researching still exploring the unknown, or learning to ask machines the right questions?
The greatest discoveries in history often arose from errors, unforeseen detours and irrational intuitions. By optimizing research to make it ever faster and more targeted, algorithmic modeling risks closing the door on the unexpected.
Can we teach the art of doubt and chance to a system designed never to be wrong?
Faced with an uninterrupted flow of computer-generated or computer-assisted papers, the academic world is running out of breath assessing and verifying what it produces. Learning the researcher's craft now demands keen discernment to separate digital noise from genuine theoretical advances.
Is the value of knowledge measured by the rigor of its demonstration, or by the human capacity to bear responsibility for it?
While public laboratories lack computing resources, the most powerful research models now belong to private tech giants. Learning cutting-edge research no longer happens only at the university, but under the control of industrial interests.
Can the pursuit of scientific truth remain a common good if its tools become patents?