In the age of AI, what should students learn most? A professor at Beijing Normal University answers: systems science.
2026-09-16
A lecture on “Why Do People Still Need to Learn?”
Last Friday, the entire faculty and student body of Zhuhai Yinghua attended a special lecture.
The speaker is Beijing Normal University. Systems Science Professor Chen Qinghua of the college delivered a talk titled “Systems Science: The Key to Decoding a Complex World.”
Seated in the audience were a group of Yinghua students preparing to study abroad—though they need not take China’s national college entrance exam, they nonetheless face their own set of examinations and an increasingly uncertain world.
Professor Chen began by sharing an observation he had made: the changes AI is bringing to education.
“What, after all, is our education teaching?”
He said he knows of a school in Funan, Anhui—a local private institution that Principal Yang has even visited. Starting in 2017, the school underwent a transformation: teachers no longer teach traditional classes; instead, AI delivers instructional videos and generates exercises, while each student uses a tablet to learn at their own pace—whether they choose to start with physics or math. The approach has proven quite effective, earning the school a strong ranking within the county.
But Professor Chen went on to say, “AI is truly remarkable, yet this very fact compels us to reflect on a crucial question.”
He asked the students in the audience, “You’re now preparing to enter the international education system and won’t be taking the domestic college entrance exam. But even if you don’t take that exam, you’ll still have to sit for some kind of test. These exams all come with questions and standardized answers. For over a decade, we’ve spent our time memorizing those answers and engaging in reflexive, conditioned‑response training—yet when we hand those same questions to an AI, it solves them far faster and with far greater accuracy. So if, after devoting so much time, effort, and money, we still can’t outperform an AI, what exactly are we teaching in our education system?”
He said that when he tackled the college entrance exam questions, he found that for many subjects, “there simply wasn’t enough time to think”—you had to know how to solve them as soon as you saw the problem. This kind of rote, drill‑and‑kill training leaves many students feeling miserable. “I bet you don’t want to suffer like that either, which is why you switched tracks,” he added.
“Questions without a standard answer are the real challenges.”
“But,” Professor Chen said, “it is precisely the problems that cannot be solved by standard answers that pose the real challenges.”
He gave several examples.
Extreme weather is becoming increasingly frequent. This year, Chengdu reached 44 degrees Celsius, and typhoon paths are growing ever more unpredictable. Why is this? No book offers a definitive answer. Nor do demographic or economic challenges come with ready-made solutions. If such answers existed, China would long ago have ceased worrying about population aging, and Japan and South Korea would have already resolved their low‑birth‑rate crises. “Even if a strategy succeeds in one country, it may not yield the same results in another.”
And then there’s human health. Everyone wants to live a little longer, but no single book offers a definitive answer to the question of “how to live to be a hundred.” There are plenty of methods online, but what works for some may be ineffective for others.
“These problems have never occurred before—how could there possibly be a standard answer?”
In the age of AI, what capabilities do humans truly need?
Professor Chen listed three points.
First, the ability to identify problems. Can we identify major issues when observing the world, and can we keenly grasp the essential core of things?
Second, the ability to discern structure. See the underlying mechanisms behind the phenomena and gain a clear understanding of how the system operates.
Third, the ability to make decisions under uncertainty. Make judgments when information is incomplete, and continuously validate and adjust as you take action.
“In fact, all these capabilities require one thing: systems thinking. You need to observe systematically, think systematically, and act systematically in order to clarify the problem and uncover the underlying logic.”
From “chain causation” to “ Feedback loop ”
Professor Chen began to explain what systems thinking is.
He said that many people are accustomed to “chain‑like causality”—A leads to B, which in turn leads to C—but the real world is rarely such a simple, straight line. In reality, the world operates through “feedback loops.” When you treat someone well, they sense it and, in turn, treat you kindly—this is positive feedback. Conversely, if you treat someone poorly, they retaliate in kind—this too is positive feedback, only negative. When body temperature deviates from 36.7°C, the body automatically regulates it back to normal—this is negative feedback.
He illustrated this with a very everyday example: “In a relationship between two people, when I see that he treats me well, I feel it and, in turn, I treat him kindly, creating a cycle. Conversely, if he treats me badly, I retaliate, which only makes him hate me even more—another cycle. Emotions, after all, operate in feedback loops.”
The whole is greater than the sum of its parts—emergence
Professor Chen also introduced a principle known as “the whole is greater than the sum of its parts”—emergence.
Ant colonies can build bridges, but a single ant cannot. The human brain contains billions of neurons, each processing only simple electrical signals, yet when interconnected, they give rise to consciousness, thought, and emotion. A single water molecule has no concept of “wetness,” but when countless water molecules come together, “wetness” emerges.
“Three cobblers with their wits combined equal Zhuge Liang,” not because each of them is exceptionally clever, but because their interactions give rise to something new.
Chaos, Self-Organization, and the Butterfly Effect
He also discussed “chaos.” In the Lorenz system, even a minuscule difference in initial conditions can lead to vastly different outcomes—this is the butterfly effect. The Three-Body Problem There is no analytical solution—not because of insufficient computational power, but because of the system’s intrinsic properties. The reason weather forecasts cannot be accurate over long time scales lies precisely in this chaotic behavior: “You cannot measure the current state of the atmosphere with infinite precision; even the slightest measurement error will inevitably lead to substantial deviations in the calculations.”
There is also “self-organization.” How do typhoons form? Without central command, they arise spontaneously through positive feedback—driven by water vapor evaporation and heat exchange—giving rise to an ordered structure. The same holds for ants building bridges: with no queen at the helm, the ants communicate via antennae and link their bodies, spontaneously forming a “living bridge.” “No one asks it to do so, nor can anyone direct all the ants to act according to their own ideas.”
Qian Xuesen’s Foresight: The Significance of Systems Science Is “No Less Than That of Relativity”
Professor Chen also cited Qian Xuesen’s assessment of systems science. In the 1980s and 1990s, Qian Xuesen remarked: “The establishment and development of systems science will spark a revolution in science and technology, one whose significance is no less than that of relativity and quantum mechanics.” Professor Chen noted that Qian Xuesen classified systems science as one of the eleven major branches of the modern scientific and technological system, viewing it as a universal framework of thought and methodology that cuts across all disciplines.
In his view, true human intelligence is not merely computational; it also encompasses intuition and emotion, requiring a synthesis of science and art. “Have you noticed that many physicists are, in fact, accomplished artists? For instance, Einstein played the violin. When Qian Xuesen was in the United States, he too was a member of an orchestra and even gave concerts. Great scientists often draw inspiration from artistic sensibilities—inspiration that does not stem solely from rational analysis.”
A Single Match: Putting Systems Science to Work
In the latter half of the lecture, Professor Chen introduced the National Youth Science Inquiry and Modeling Competition—specifically its Systems Science track. This competition is open to primary and secondary school students nationwide and consists of three stages: a preliminary round, a semi-final round, and a national final. The preliminary round involves answering multiple-choice questions on a computer; the semi-final requires selecting one project from a pool of 23 to develop; and in the national final, participants further refine their work through an on-site exhibition and defense.
Group C1 (Introductory Exploration)
A question is called “ Droplet Electrostatic Generator “Using soda cans, plastic bottles, and metal wire to build a device that makes water droplets produce electrical sparks—‘If you have the time, you can try building one; it allows the electric charge between two metal cups to gradually build up, eventually resulting in a discharge.’”
There’s also a question called “The Bowl of Knowledge,” which prompts students to consider how to review most efficiently—balancing time between new and old material to determine the optimal ratio that maximizes what the brain retains.
Group C2 (General High School Category)
“Who is the key figure in the family?” — Have students construct a family network and determine who wields the greatest influence. “How do bird flocks perform their spectacular dances?” — Simulate flocking behavior using three simple rules. “The staircase from order to chaos” — Use… Logistic map Draw a period-doubling bifurcation diagram and witness firsthand how chaos emerges. “Chaos Philosophy in Dough Kneading” — through a physical experiment of stretching and folding, gain an understanding of the mathematical principles underlying mixing efficiency. “Ant Colony Algorithm Lab” — use the evaporation and accumulation of pheromones to simulate how collective intelligence finds the optimal path.
C3 Group (High School Elite Group)
“The Avalanche of Sand Piles” — investigating how tiny perturbations can trigger large-scale collapses, and verifying power-law distributions. “Rumors Travel Faster Than the Truth” — simulating how rumors spread through networks to identify critical nodes that can halt their propagation. “Why Do Ghost Traffic Jams Occur?” — using cellular automata to model traffic flow, exploring how an inadvertent brake can spark a congestion several kilometers away. “How Columns Can Save Lives” — examining how seemingly obstructive columns in crowd evacuations can, paradoxically, enhance overall evacuation efficiency. “A Stunning Leap from Insulation to Conductivity” — studying percolation: when metal and glass beads are mixed in a certain proportion, the system suddenly becomes conductive. “The Butterfly Effect in Chaotic Pendulums” — constructing a double-pendulum apparatus to observe how minute differences in initial conditions get exponentially amplified. “Can Global Affairs Be Quantified?” — using United Nations voting data to quantify international relations and delineate geopolitical blocs.
“There is no single correct answer to any of the questions,” said Professor Chen. “Each one requires observing phenomena, abstracting underlying mechanisms, constructing models, and validating and refining them.”
“This is the uniquely human capacity for thought and creativity.”
Toward the end of the lecture, Professor Chen said the following:
“AI can perform ‘from A to B’ reasoning, but when it comes to navigating a complex world, redefining problems, delineating their boundaries, and exploring solutions—these are capacities of thought and creativity that remain uniquely human.”
The teachers and students of Yinghua sat in the audience, having listened to the entire lecture. They may not yet have fully grasped the full implications of terms like “positive feedback,” “emergence,” and “chaos.” But there was one remark by Professor Chen that ought to be remembered—
“Schools spend a great deal of time having students memorize standard answers. But since the future hasn’t arrived yet, how can you possibly know what kinds of problems you’ll face? If the questions themselves remain unclear, where would the so‑called standard answers come from? And why should there be a single, fixed answer for something that has never happened before?”
There are no standard answers in this world. And learning to think and act in a world without definitive answers—that is the true purpose of education.
Previous Page: