Shanghai, China — A new article by University of Kansas scholar Yong Zhao argues that education needs not only a new response to artificial intelligence, but a new theory of how schools can change. In "Obsolete Schooling and the Courageous Minority: Rethinking Educational Change in the Age of Artificial Intelligence," published in ECNU Review of Education , Zhao contends that prevailing approaches to reform are poorly suited to the scale and urgency of the challenge.
The rapid development of AI has made this challenge impossible to ignore. Machines can now perform many tasks traditionally assigned to students: generating essays, answering routine questions, summarizing texts, and producing acceptable schoolwork. But Zhao's central claim is that the significance of AI is not primarily technological. AI is exposing an educational model whose routines and measures of success have become increasingly obsolete.
The usual responses—banning tools, tightening rules, detecting misuse, or adding AI guidance to existing programs—do not address the deeper problem, Zhao argues. Schools must undertake transformational change: reconsidering what is worth learning, how learning should be organized, and what purposes school should serve. Yet conventional theories of school reform have rarely produced that kind of change.
For decades, reform has been organized around large-scale policy initiatives, comprehensive frameworks, and system-wide implementation. These efforts typically aim for coherence and uniformity: a common reform agenda, common expectations, and common measures of success. But schools have repeatedly absorbed such initiatives into their existing routines, preserving age-graded classrooms, standardized curricula, subject divisions, rankings, and uniform assessments. Reforms may change language, techniques, or technologies, but seldom change the underlying grammar of schooling.
Zhao explains this persistence through the idea of school as a "peace treaty." Schools are not merely educational organizations. They are also settlements among students, families, teachers, school leaders, universities, policymakers, and communities, whose interests and expectations do not fully align. Tests, grades, schedules, credentials, and familiar instructional routines help maintain this settlement. A reform that attempts to change everything for everyone at once threatens too many interests and therefore predictably produces resistance, symbolic compliance, or dilution.
The article proposes the courageous minority as an alternative theory of educational change. Rather than waiting for a system-wide mandate or universal consensus, a relatively small group of teachers, students, leaders, families, and community partners can create meaningful alternatives in the spaces they actually control. A classroom, advisory, exhibition, capstone, interdisciplinary project, community partnership, or school-within-a-school can become a protected space in which different assumptions about learning are put into practice.
This approach does not treat small-scale action as a second-best substitute for "real" reform. Drawing on panarchy theory, Zhao argues that complex systems do not change only from the top down. Smaller, semi-protected parts of a system can experiment, develop viable alternatives, attract allies, and influence larger structures over time. A courageous minority can thus generate evidence, new norms, and examples of possibility before a larger system is ready to authorize them.
The theory also recognizes urgency. Students are already living with AI, navigating assignments that may have little meaning beyond compliance, and encountering assessments that reward performances machines can increasingly produce. Even a rapid policy response would arrive too late for many students now moving through school. The courageous minority approach asks educators not to wait for permission to begin making learning more personalized, more problem-centered, and more connected to human contribution.
For school leaders, this means creating and protecting room for experimentation rather than imposing another uniform reform. For teachers, it means designing learning around meaningful problems, student agency, long-term inquiry, revision, and public contribution. For students and community partners, it means becoming co-authors of educational change rather than recipients of it.
The article's argument is already being tested through a global network of about 20 schools. Participants from different countries are designing locally grounded experiments, including learner-agency days, capstone projects, student podcasts, AI-supported inquiry, problem-finding initiatives, innovation groups, and schools within schools. Their work illustrates the article's central proposition: transformational change need not begin when an entire system agrees. It can begin when courageous people act in the zones of possibility they already possess.
In the age of AI, Zhao concludes, the future of education may depend less on the newest technology than on whether enough people are willing to create forms of schooling that students find genuinely worth doing.
