AI News Feed
Market watch
Products & Applications

Nanjing professor’s new course rule: CS students without tokens should drop out

A Nanjing University associate professor sparked heated debate by saying computer science students without paid AI tokens should drop out — and requiring his students to self-fund tokens and avoid “ancient programming” in class.

Jiang made the comment while discussing collective anxiety among computer science students in the AI era. He described students using AI tools like Doubao to finish homework while receiving no actual training, and another group who studied diligently for four years only to find that AI can already do what they learned. He also cited so-called “perfect resumes” that list papers and awards but lack GitHub links, personal pages, or preprints, making applicants indistinguishable.

Jiang said the course aims to ease such anxiety by bringing back the fun of “vibe coding.” He stressed that students must learn not only to write code, but to break down tasks for AI, choose models, control costs, and verify outputs. He recommended the cost-effective DeepSeek-v4-flash model, suggested students recharge 100 yuan on DeepSeek, and offered help to those facing token shortages.

The remarks met with significant support online. One Zhihu user said Jiang’s statement was “too conservative,” suggesting universities cancel all classes and buy tokens for students instead. Another called it “far better than many empty courses.” Some compared tokens to lab reagents in chemistry, arguing that universities — not students — should bear such costs.

Jiang Yanyan received his bachelor’s and doctoral degrees from Nanjing University in 2011 and 2017, won the CCF Outstanding Doctoral Dissertation Award, and received best paper awards at ICSE 2021 and SOSP 2023. His research focuses on system software and software automation, and he has recently shifted toward agentic AI. He publicly shares his courses on Bilibili, where his lecture videos have accumulated more than five million views.

Jiang is known for an unconventional teaching style, maintaining a Linux-terminal-inspired homepage and frequently using internet memes in class. His new course continues that approach, with rules like “ancient programming is forbidden” that have become the latest topic of campus conversation.