As the new academic year begins, presidents from leading universities including Peking University, Fudan University, and SUSTech are sounding a unified alarm about artificial intelligence. In their opening ceremony addresses, they are using their first lecture to urge the incoming class, described as the "AI-native generation," to avoid over-reliance on AI and guard against "cognitive unloading," "illusory mastery," and "outsourced thinking."
Asking AI for answers has become a default habit. A survey of 1,953 undergraduate students across more than 400 universities nationwide reveals that 99.18% of respondents use AI. Students are turning to AI for everything from essay angles, research materials for reports, and paper frameworks, demonstrating that AI is deeply embedded across the entire chain of university learning, from problem formulation and logical argumentation to knowledge retrieval and even thought processes.
More concerning is that AI's constant presence is threatening to steal the spotlight from genuine learning. Teachers lament that they can no longer judge whether a student has mastered a subject based on coursework completion. A well-structured paper or a thoughtful reading reflection might be produced by nothing more than a student's conversation with a chatbot. While efficiency improves, the parts that are skipped, such as reading original texts, sorting out ideas, and even the seemingly tedious task of data analysis, represent essential foundational thinking training. If the only goals are for students to hand in assignments and teachers to complete teaching, without achieving growth in ability and cognition, it amounts to an "educational stall" that merely burns computational tokens.
AI is not a monster, but the concerns are not unfounded either. We must acknowledge that if humans are viewed purely as vessels of knowledge, we are no match for AI. Its massive databases and abilities in information retrieval, storage, processing, integration, and data-driven computation and reasoning have long surpassed human levels. AI has also broken down the boundaries of higher education. Knowledge that might take a semester to deliver in a lecture hall can now be dispensed by AI in minutes, even offering judgments based on certain generative logic.
As AI becomes increasingly intelligent, is there still a need for humans to learn, and where does the space for education lie? If humans choose to "lie flat" and surrender their thinking abilities, the only outcome would be that machines become more and more like humans while humans become more and more like machines. So, where does the real difference lie in skills? From many practical examples, AI can provide answers but may not be able to pose genuinely valuable questions. AI can list information but struggles to make value judgments. AI can generate solutions but cannot bear the responsibility for choices. In short, the core advantage of humans is not about memorizing more or calculating faster than machines, but about the ability to ask "why," judge what "should" be done, imagine "what else could be," and then put ideas into practice, test them in the real world, and make continuous corrections.
Ultimately, the true impact of AI is not about how many tasks it can complete, but that it forces humans to think about what is worth researching and what constitutes knowledge. The key to the future is not about declaring a winner between man and machine, but about exploring how to establish a new connection between the two. This challenges universities to reassess their educational models. The old logic of "knowledge delivery – assignment evaluation – exam assessment" is no longer viable. To upgrade the educational model, we should guide students to transform from passive receivers of knowledge into problem posers, meaning constructors, and value guardians. Teachers should shift from being one-way transmitters of knowledge to guides of thinking, organizers of dialogue, and supporters of creativity. Making technology serve human growth, rather than replace it, is both the main principle and the bottom line.
"The way of the great learning involves manifesting virtue, renovating the people, and abiding by the highest good." In the age of AI, revisiting this classic teaching offers fresh insights.