Universities must add and subtract curricula, shift from transmitting knowledge to teaching students how to learn, and redefine their infrastructure around computing power rather than libraries, said academics and industry leaders at a forum, as artificial intelligence upends what constitutes talent in the modern era.
The HKU Talent Forum, recently held at the University of Hong Kong Beijing Center, gathered scholars from universities and research institutions — alongside executives from shopping platform JD.com and tech giant Xiaomi — to discuss talent cultivation, innovation and the mission of universities amid technological change.
Lin Chen, vice-president of HKU, said AI is both an opportunity and a challenge. Lin said outstanding talent must possess an inquisitive mindset, a solid theoretical foundation and the capacity for independent, innovative judgment, and must be able to make responsible choices in the face of uncertainty.
"We hope to respond to the times with innovation and connect the world with openness," Lin said, calling for the cultivation of responsible talent with a global outlook and a sense of national responsibility.
Tan Jianrong, an academician of the Chinese Academy of Engineering, said that universities must both break down disciplinary barriers to foster interdisciplinary talent and — given students' limited bandwidth — shift from imparting knowledge to cultivating the ability to acquire it.
"We must teach fishing, not just give out fish," Tan said.
Xue Lan, dean of Tsinghua University's Schwarzman College, offered a measured assessment of the fast-evolving technology's current reach, saying that large language models have had significant impact in narrow domains such as coding, but their practical influence in healthcare, education, and autonomous driving remains limited and still falls short of expectations.
Still, Xue warned that AI has shattered higher education's monopoly as the sole gateway to professional knowledge, and predicted that future universities will be defined not merely by gymnasiums or libraries, but also by computing platforms, data centers and intelligent tutoring systems.
His remarks come as Chinese universities are quietly reclassifying AI access from a scarce research resource into basic campus infrastructure. The University of Science and Technology of China launched its "Token Program" mid-year, granting each student 100 million tokens per day. In September, Beijing University of Posts and Telecommunications opened free computing accounts for nearly 10,000 freshmen. On Oct 26, Tongji University began issuing 100 million tokens per week to all faculty and students.
Industry executives underscored the shifting demands on graduates.
Duan Nan, vice-president of JD.com, described an industry that iterates "on a daily basis", requiring strong execution and interdisciplinary agility.
"In an era of human-AI symbiosis, the critical skill is judgment — the ability to evaluate the quality of AI output and to ask the right questions," Duan said.
Wang Bin, director of Xiaomi's AI Lab, offered a candid account of how uncertainty has reshaped corporate strategy.
"From the top of the company down, including myself, we are all anxious about how to respond to the future society," Wang said.
The first essential response, he argued, is embracing change itself — lab structures and organizational arrangements at Xiaomi are in constant flux, and resisting such a phenomenon is no longer a viable option.
The second is resilience and relentless learning. Wang admitted that his own Natural Language Processing (NLP) expertise has become largely obsolete, and he now uses new AI tools daily in his work. Xiaomi has responded institutionally by running "AI-native" experiments across the company through pioneering teams, sharing lessons on a weekly and even daily basis.
masi@chinadaily.com.cn
