In the age of artificial intelligence, reforming math education in Korea isn’t just a good idea — it’s long overdue. AI runs on mathematics. Not just any math, but the kind that quietly powers how machines “think,” recognize patterns, and make decisions. One of AI’s core ideas is similarity — figuring out how close or far apart things are. To quantify this, early machine learning models leaned on fundamental mathematical concepts: spatial relationships, distances, and proximity. Optimization lies at the heart of it all. Gradient descent, for instance, is essentially calculus on autopilot — repeatedly adjusting parameters to improve predictions. Vectors and matrices are the unsung heroes, managing high-dimensional complexity and enabling clever shortcuts like the kernel trick, which produces results without computing every transformation directly. Strip away the code and jargon, and you’ll find what I often argue: AI is really just mathematics in disguise, running the show behind the scenes.
Recognizing this connection, Korean educators made a bold move by introducing a new high school subject: “AI Mathematics.” It got people talking. My international peers (mostly professors) began requesting copies of the textbook. Even more surprising, adults who had once sworn off math started returning to evening math classes and AI boot camps. Their reflections are strikingly consistent: The math they learned in school — crammed, rushed, and painfully test-driven — didn’t prepare them for solving real-world problems. Now, approaching it from a fresh angle, they realize math is unavoidable — and that they need to rebuild from the ground up.
This renewed reckoning with mathematics stands in sharp contrast to the rigidity of Korea’s current education system. Recent curriculum reforms, driven by well-meaning liberal politics, removed essential topics like vectors and matrices from the high school syllabus, labeling them “too difficult.” But let’s be honest: When students are expected to solve challenging problems in under five minutes, everything becomes hard. The issue isn’t difficulty — it’s how we teach and test. We’ve turned math into a speed contest instead of a chance to think deeply and learn with intention. Students focus on memorizing shortcuts, not understanding ideas or applying them flexibly. So when they reach college and face new material — challenging or not — they panic, hunt for answer keys, and often get lost. Some even drop out.
The truth is, our formal system keeps dropping the ball. High school math remains a test-prep machine. And in college, there's little distinction between math that serves general learners or applied fields, and math intended for future mathematicians. We have non-majors wrestling with abstract algebra designed for graduate-level coursework. Even within math departments, many students aren’t aiming to become researchers. Some plan to teach at hagwons; others are planning an exit. Whatever the case, we’ve lost the soul of mathematical practice: reading and writing math, thinking aloud with peers, engaging with its history and culture, and exploring real, open-ended problems.
So what now? We need a few major changes. First, college entrance exams should reflect the kinds of mathematics actually used in different academic and professional fields. No more one-size-fits-all nightmare tests. Second, we should offer full national scholarships to students who demonstrate not only talent, but also genuine intellectual curiosity in math and in other foundational disciplines — like history and philosophy — that may not promise high salaries but enrich the moral and intellectual fabric of civilization. Third, we need to overhaul how we certify math teachers. Current exams are heavy on abstract theory and formal analysis, and light on what truly matters — fieldwork, real-world applications, and teaching practices that bridge college-level math with school mathematics.
We also need to give our math teachers more support. The curriculum itself isn’t bad — it actually covers a lot of important ground. The real problem is the intense focus on getting into top colleges. That pressure often hijacks how math is taught and learned. One persistent challenge is the lack of curricular coherence across elementary, middle, and high school, which makes it hard to build consistent learning progressions. While some advocate for incorporating modern applications like AI and real-world problem solving, most programs still revolve around traditional topics like calculus and analysis. The bigger issue is that people can’t agree on what math is for — mathematics as a test-taking skill, a language for science, math for its own sake, or a form of cultural literacy. That confusion puts a lot of pressure on teachers and makes it harder to focus on real learning. So, teacher education programs should aim to help future educators not only build strong instructional skills but also reflect critically on the broader role of mathematics in society.
This is just the beginning. But if we start here, we might finally turn math into what it ought to be — not a source of stress, fear, or a shortcut to top college admissions, but a source of wonder. A discipline that drives and deepens science. A deeply human way to understand and shape the world — especially now, in a time when AI touches nearly everything we do.
Lim Woong
Lim Woong is a professor at the Graduate School of Education at Yonsei University in Seoul. The views expressed here are the writer’s own. — Ed.
khnews@heraldcorp.com
