For centuries, schools have operated on a simple assumption: knowledge is scarce, teachers possess it and students must acquire it. Schools have built elaborate systems of curriculum, instruction, testing and credentials around this assumption. Artificial intelligence is making that assumption increasingly difficult to defend.
Today, students can obtain explanations, generate essays, solve mathematical problems and explore complex questions within seconds. Tasks that once required considerable learning can now be accomplished, at least superficially, with a few prompts.
This raises two fundamental questions: What does learning mean when answers are readily available? And does easy access to knowledge make us smarter or intellectually weaker?
My answer is that AI does not determine the future of learning. Our conception of learning determines whether AI becomes an instrument of intellectual growth or intellectual decline.
The fundamental problem is not that AI has made learning unnecessary. It is that AI has exposed the limitations of what schools have traditionally considered learning.
Knowing the answer to a question is not the same as understanding a problem. Producing an essay is not the same as having something meaningful to say. Solving a mathematical equation is not the same as knowing when and why mathematics matters.
I believe schools must embrace three fundamental shifts in the age of AI.
The first step must be moving from standardized learning to personalizable learning.
Every child is a unique human being with different strengths, interests, passions and potentials. Yet schools have traditionally expected everyone to learn largely the same content and demonstrate success against the same standards.
AI offers opportunities to support a different kind of education: helping each student discover who they are, what they care about, what they are good at and what they might become.
This does not mean abandoning foundational competencies. All students need a reasonable foundation for participating in society. Beyond that foundation, education should enable different learners to pursue different forms of excellence.
The second step must be shifting from answering questions to finding and solving meaningful problems.
Schools have traditionally rewarded students for answering questions posed by someone else. AI is exceptionally good at answering many such questions.
The more important human capability is increasingly the ability to notice problems worth addressing, formulate meaningful questions, exercise judgment and create solutions that matter to others.
Instead of asking students to write an essay about environmental problems, why not ask them to identify a local environmental challenge and develop a workable response? AI can help them investigate, analyze and design. But students must determine what matters, whose needs deserve attention and whether their solutions actually work.
The third step must be moving from individual competition to human interdependence.
Traditional education treats learning largely as an individual accomplishment. Students compete for grades, rankings and credentials.
But human beings thrive through cooperation. Schools should help students contribute their unique talents, appreciate differences, collaborate with others and create value collectively.
The objective is not simply to produce more knowledgeable or employable individuals. It is to cultivate human beings capable of building a better, more peaceful and more prosperous society.
The concern that AI will weaken effort, memory and critical thinking is legitimate. But it is not new.
Similar concerns accompanied writing, printed books, calculators, computers and the internet. Every technology that reduces the effort required to accomplish something also changes what people practice and what they may lose. AI is no exception.
If students use AI to complete assignments without thinking, they may indeed learn less. If they rely on AI for every answer, they may fail to develop essential knowledge and judgment. But the opposite is equally possible.
AI can help students explore questions beyond their existing knowledge, receive feedback, consider alternative perspectives, test ideas and pursue interests that traditional classrooms cannot easily accommodate.
The difference lies in what students are trying to accomplish and how they engage with AI.
Consider two students using AI to write an essay.
One asks AI to produce the essay, makes minor adjustments and submits it. The student has completed an assignment but may have learned very little.
The other has a question they genuinely care about. They use AI to explore evidence, challenge assumptions, examine competing arguments and improve their writing. They remain responsible for the argument and its consequences.
Both students used AI. But their learning experiences were profoundly different.
We must also reconsider effort and memory. Effort is not inherently educational. Struggle contributes to learning when it serves a meaningful purpose. Likewise, students need sufficient knowledge to understand their world and exercise judgment, but memorizing everything is neither possible nor desirable.
Critical thinking develops through questioning, evaluating, experimenting and reflecting, not simply through restricting access to information.
AI can weaken thinking when it replaces intellectual engagement. It can deepen understanding when it enables more meaningful intellectual engagement.
These possibilities are guiding two initiatives I am developing with colleagues and educators.
The first is YEE AI, an AI-supported learning platform designed around personalizable learning, problem finding and solving, and human interdependence.
Rather than delivering prescribed content more efficiently, YEE AI aims to help students discover their strengths and interests, identify meaningful problems, collaborate with others, and document their learning and contributions. AI should expand students' agency and creativity, not become another mechanism for controlling their learning.
The second initiative is Courageous Minority, an international community of teachers and school leaders committed to transforming education within the spaces they can influence.
Educational transformation cannot wait for entire systems to change. Teachers can begin by redesigning learning experiences in their own classrooms, giving students greater agency, encouraging authentic problem-solving, and using AI to expand human possibilities.
The two initiatives complement each other: YEE AI explores how technology can support a new educational paradigm, while Courageous Minority explores how educators can bring that paradigm to life in real classrooms and schools.
Ultimately, AI invites us to revisit a question schooling has avoided for too long: Are we educating children to perform tasks, or are we helping them become capable human beings who can live meaningful lives and contribute to the world?
AI makes many traditional educational tasks easier. It does not make becoming human any easier.
