Human Regression Through Cognitive Delegation to AI: An Examination of Risks Associated with Student AI Usage in the Post-ChatGPT Era
The act of humans voluntarily delegating thought and decision-making to AI poses a serious risk that could lead to humanity’s intellectual regression. This could trigger a chain reaction that lowers average intelligence within just a few generations, making it more serious than the problem of recommendation algorithms. The danger is particularly acute because, unlike something being “taken away,” what is “given away” cannot be reversed. Chatbot AI structurally embodies the problem of hallucinations (bullshit), which is a fundamental limitation that cannot be resolved even with techniques like RAG (Retrieval-Augmented Generation). Therefore, chatbot AI is an entirely unreliable tool. Students who use AI like a calculator outsource cognitive effort, leading to a decline in their ability to think for themselves—that is, their critical thinking skills. According to MIT research, students who used ChatGPT showed lower brain activity, relied on copy-pasting, and subsequently were unable to write properly without AI. This voluntary dependence causes human capabilities to atrophy (de-skilling), and ultimately leads to a pessimistic prospect of entering a “co-evolutionary circuit” where the majority of humans regress intellectually while only AI capabilities improve. The ability to think, like physical exercise, can only be maintained through repetitive training. Therefore, humans must preserve their uniqueness as “first movers” who challenge novelty, and must continue mental training relentlessly without surrendering originality in rage against machines.
Keywords: Intellectual Regression, Voluntary Delegation, De-skilling, Critical Thinking, Co-evolutionary Circuit
주제어: 지적 퇴보, 자발적 위탁, 탈숙련화, 비판적 사고, 공진화 회로