BEIJING, China– In a lecture hall at Shanghai’s Fudan University, the traditional classroom is being turned upside down. To ace the final exam in a data mining course, students aren’t solving algorithms—they are tasked with writing original, complex questions designed to challenge advanced AI models like Claude and DeepSeek .
The harder the AI struggles to answer, the higher the student’s grade, according to a report published by Fudan University.
“Traditional testing methods are facing new challenges in the age of AI,” explains Professor Xiao Yanghua, who teaches the course. “If we ask students to solve a standard algorithm, AI will always be faster. We must move toward training them to judge, direct, and identify the structural flaws of these models.”
As educators worldwide debate the role of artificial intelligence, China is rapidly shifting from experimentation to a massive, systemic rollout. Under its national “AI Plus Education” roadmap, the country aims to build a comprehensive AI-driven education system across all learning levels by 2030.
The objective is clear: China is not just teaching students how to use AI, but how to understand, critique, and co-exist with it, preparing them to navigate both its profound merits and inherent challenges.
Empowering classrooms from K-12 to vocational education
Across China’s vast educational landscape, AI is moving from a passive subject of study to an active partner in the learning process.
At Fendou Primary School in Beijing, teachers use AI-generated, interactive 3D models to help students rotate and disassemble geometric shapes on-screen, transforming abstract math into a hands-on experience.
Beijing No. 13 High School utilizes AI agents to track students’ digital learning footprints, creating “learning health reports” that allow teachers to replace generic homework with targeted, individualized assignments.
At Beijing No. 35 Middle School, an AI-powered code reviewer diagnoses student programming errors in under two minutes, offloading hours of grading, so teachers can focus on high-value, one-on-one coaching.
In vocational education, Shandong College of Tourism and Hospitality uses “Baize,” an AI digital assistant, to simulate real-world hotel check-ins and guest services, letting students practice working alongside AI in virtual environments.
Navigating the friction and imperfections of AI
Chinese educators openly acknowledge that AI is a powerful but highly imperfect tool. While algorithms excel at pattern recognition, they still struggle with human-centric capabilities like evaluating creative expression, understanding nuanced reasoning, and protecting data privacy.
To safely manage these friction points, China is building a robust ecosystem around the technology: AI literacy is being integrated into national teacher qualification exams, shifting an educator’s role from a mere technology user to a “digital leader” who guides the moral and creative development of students.
To prevent resource disparity and data silos, Beijing’s Xicheng District connects 90 pilot schools with top universities and tech firms to safely test and share AI teaching tools.
Cultivating intellectual agility for an uncertain future
Ultimately, China’s educational shift is about preparing minds for an unpredictable job market. At an April press conference, Fudan University President Jin Li argues that the ideal graduate in the AI era must be like a “stem cell”—possessing the multi-directional potential to cross disciplines and continuously self-update as technical skills become obsolete.
Classrooms are no longer just places of rote instruction; they have become laboratories for human-machine collaboration. By teaching students to judge, direct, and find the structural flaws of AI systems, China is ensuring that human wisdom remains the ultimate compass directing where technology goes next.
Author: Wang Yangyang and Liu Yushan
