Artificial Intelligence has transformed access to information, but it has also created an illusion of knowledge. This article reflects on the evolving nature of learning in the AI era and the shifting priorities in higher education under India’s National Education Policy (NEP). It warns against overemphasizing skills and tools at the expense of theory and critical inquiry, arguing for a balanced model of education that fosters both understanding and innovation. When we were children, our learning journey was grounded in books written by experts and scholars who had devoted their lives to exploring their disciplines. Those texts not only provided information, but also shaped our curiosity, taught us to think critically, and rooted our understanding in theory. The depth of such engagement built intellectual resilience. Today, the learning landscape has changed profoundly. Artificial Intelligence (AI) systems such as ChatGPT can generate instant explanations, essays, or even computer code. The convenience is undeniable, yet it comes with an invisible cost and that is the illusion of understanding. Learners often accept AI-generated responses without grasping the principles behind them. When they ask a question and receive an answer, many feel content, believing they have truly understood, but they have only just scratched the surface. Stephen Hawking once remarked that “the greatest enemy of knowledge is not ignorance; it is the illusion of knowledge.” In the AI age, this illusion has become pervasive. Ignorance at least provokes curiosity but the illusion of knowledge ends inquiry. When students begin to trust the output of a machine without reflection or verification, learning turns mechanical. It breeds false confidence and intellectual stagnation. Having taught Computer Science for more than a decade, I have seen this shift firsthand. Earlier, students would spend hours understanding algorithms, debugging programs, and exploring the logic behind every line of code. Today, a single AI-generated script may complete that task in seconds. As we get faster and more productive, the depth of our understanding quietly slips away. The foundational cognitive skills like reasoning, abstraction, and problem-solving that Computer Science education seeks to nurture are quietly eroding. India’s National Education Policy (NEP) 2020 rightly emphasizes interdisciplinary, flexibility, and employability. These reforms are necessary, but their partial or superficial implementation can have unintended consequences. If colleges and universities, which are the highest seats of learning, focus primarily on skill training while neglecting theory, they risk producing students who become tool users rather than creators of knowledge. Skills evolve as tools become obsolete, but theories endure over time. Knowing how to operate a neural network is not the same as understanding how learning occurs within it. Yet in many classrooms, the focus has shifted toward “what works” instead of “why it works.” This trend, if unchecked, will produce graduates adept at using technology but unprepared to question or improve it. The deeper danger lies in a growing global divide between those who create and control technological tools and those who merely consume them. The former group designs algorithms and shapes the future; the latter depends on systems they do not understand. Education that limits itself to tool usage contributes to this imbalance, leading to a subtle form of digital dependency where technology commands, and humanity complies. To reverse this trend, educators and learners alike must rediscover the essence of inquiry. AI should be seen as a partner in exploration, not a substitute for thinking. True education demands humility i.e. the awareness that every answer opens more questions. It is through reflection and critical engagement that information matures into knowledge. To safeguard deep learning in the AI era, we must:

1. Approach AI outputs with scepticism and intellectual humility.
2. Use AI as a guide to explore, not as the final source of truth.
3. Continue engaging with original books, research, and expert discourse.
4. Encourage conceptual understanding before implementation.
5. Preserve curiosity that sparks discovery.

Education must remain a process of awakening the mind, not automating it. The goal is not to produce individuals who merely know how to use tools, but those who understand, question, and create them. The future of learning will depend not on how efficiently we access information, but on how deeply we can comprehend it. The challenge before us, then, is both simple and profound; are we educating thinkers or merely training tool users?   (The Author is an Assistant Professor of Computer Science in the School of Computing at the Samarkand International University of Technology, Samarkand, Uzbekistan. With over a decade of teaching and research experience, his academic interests include digital image forensics, machine learning, computer vision, cyber security, and Blockchain. He writes on the philosophy of technology and the evolving nature of education in the AI era)