Series Index
1. Why Learning Standards Are Breaking Down in the AI Era
How AI is eroding traditional learning benchmarks and blurring the question of what students should learn.
2. What Is Language Learning For Now?
Why language is shifting from a tool of expression to one of interpretation and understanding.
3. What Does It Really Mean to Be Good at English?
A redefinition of English proficiency around structure, context, comprehension, and literacy.
4. More Dangerous Than Dopamine Addiction: The “Click” Brain
How AI’s response structure is conditioning children to react before they think.
5. In the AI Era, By What Standard Should We Learn?
A concluding framework for educational choices that connects exams, judgment, and the future.
By Kang Sang-woo, Head of Future Strategy, CREVERSE
“Why think about it? You can just ask AI and get the answer instantly.”
This kind of response is no longer unusual when talking with students. Faced with a difficult problem, they open AI before they even begin thinking. When a long passage appears, instead of rereading it, they ask, “Can you summarize it in one sentence?” If a conclusion does not appear within 30 seconds, they move on. When they encounter something difficult to understand, they no longer struggle through it themselves—they immediately look outside for the answer.
We often describe this as “dopamine addiction” or short-form content addiction. That is not entirely wrong. But in actual classrooms, educators are sensing something much heavier. The real issue is not simply a habit of consuming content quickly. It is that the thinking process itself is disappearing.
Recently, researchers at the University of Pennsylvania’s Wharton School gave this phenomenon a precise name: cognitive surrender. The moment people encounter an AI-generated answer, they stop making the effort to think critically and begin accepting the result without verification. What is even more striking is that the more capable the AI appears, the stronger this tendency becomes. In experiments where researchers adjusted AI reliability levels, participants using higher-performing AI were actually less likely to detect critical errors. The moment verifying the AI’s answer feels cognitively more expensive than simply trusting it, people give up on verification altogether. Researchers explained that this tendency stems from humanity’s instinct to conserve mental energy.
To understand this more deeply, we need to revisit psychologist Daniel Kahneman’s framework of human cognition. Kahneman divided thinking into two systems. System 1 reacts quickly and intuitively, while System 2 thinks slowly, deeply, and analytically. Normally, humans respond first through System 1 and then move into System 2 when deeper reasoning is required. Rereading a confusing sentence, connecting logic across paragraphs, inferring meaning, and reconstructing understanding are all functions of System 2.
The problem is that today’s AI environment is rapidly weakening this transition process. Children are not failing to think; they are constantly reacting. But those reactions end quickly before they can develop into deeper reasoning. Even the immediate responses of System 1 are increasingly shaped by a form of “implicit agreement,” to the point that System 1 itself becomes paralyzed, repeatedly reinforcing a cognitive pattern that never progresses into System 2 thinking.
The signs are visible in classrooms. When asked a question, students pause briefly as if thinking, and then suddenly stop. It is not necessarily fear of being wrong. Often, the answer is within reach if they think a little longer. But they increasingly struggle with the act of walking that distance themselves. Brains repeatedly exposed to environments where answers arrive instantly become optimized for a simple assumption: “If I wait a moment, someone—or something—will tell me.” This is not simple laziness. It is a warning sign that the ability to hold onto a problem, connect ideas independently, and sustain thought until the end is slowly disappearing.
I call this phenomenon the Click-Response Brain, or simply the Click Brain.
The Click Brain is extremely good at reacting quickly to external stimuli. One search produces an instant summary. One AI prompt generates a polished sentence or completed answer. Results appear before struggle even begins. As this convenience repeats itself, children’s brains become increasingly accustomed to skipping the process itself. This is not merely an issue of attitude or willpower. The deeper problem is that the brain’s operating structure is being reorganized away from “thinking” and toward pure “reaction.”
Human learning was originally inefficient. We reread difficult sentences, connect ideas across passages, infer context, and gradually construct meaning. Sometimes we fail. Sometimes it takes a long time. Sometimes it feels frustrating. But it is precisely within that slow and uncomfortable process—the full activation of System 2—that thinking ability grows. Short-form media shortened the length of attention. AI is now beginning to reduce the need for thinking itself. Conclusions come first, while process becomes compressed. As “consuming answers” replaces “experiencing thought,” the Click Brain increasingly struggles to follow long passages, infer relationships between sentences, and reconstruct the writer’s intention.
This shift is now appearing clearly in academic outcomes as well. In South Korea’s 2025 College Scholastic Ability Test (CSAT), the proportion of students achieving the highest grade in English fell to 3.11%, the lowest level since the introduction of absolute grading. More importantly, similar signals are appearing even among students far removed from the CSAT itself. In the March 2026 nationwide academic assessment for first-year high school students, the cumulative percentage of students achieving Grade 1 in English was just 4.38%—roughly half the average range of 8–9% recorded over the previous four years. Grade 2 cumulative rates fell from 19.2% to 12.83%, while Grade 3 rates dropped from 32.1% to 26.39%. This is no longer just a problem affecting top-performing students. The entire middle layer is weakening.
This is not simply a matter of exam difficulty. Modern English assessments are no longer measuring “how much students memorized.” Increasingly, they are testing how long students can sustain thought. Exams demand the endurance of System 2, while the Click Brain is becoming progressively optimized for short reactions. The growing gap between those two realities is now appearing directly in student performance.
Ultimately, we must return to the importance of process. AI already generates sentences faster and more naturally than humans. Fluency and accuracy alone are no longer enough to define competitiveness. What matters increasingly is the structure of thinking: the ability to read, connect, infer, and reconstruct ideas in one’s own language. In the age of AI, the central question is no longer simply what you know, but how you understand. Not how quickly you respond, but how deeply you can think.
The habit of skipping the thinking process initially appears efficient. But over time, it dramatically weakens children’s cognitive endurance. By contrast, students who resist the temptation to instantly “click,” who continue reading until the end, connect ideas independently, and struggle through problems with their own reasoning, gradually create a widening gap between themselves and others. That is the true capability now being most clearly revealed through English literacy.
In an era where technology can generate answers for us instantly, what kind of learning should remain most valuable for our children?
The answer is not simple. But the standard is becoming clearer.
In the next episode, we will continue this discussion by exploring what principles should guide learning in an age where AI increasingly provides the answers for us—and what we must ultimately leave behind for the next generation.
About the Author
Kang Sang-woo holds a master’s degree in business administration from Seoul National University, studied economics at Duke University, and graduated from Daewon Foreign Language High School. He is currently Executive Director of Future Strategy at Creverse, where he leads content and curriculum for THE OPEN. His previous experience includes heading strategy at AI companies in the medical and entertainment sectors, as well as working as a U.S. college admissions consultant.
