Article from The Economist (UK), March 25
Headline: China Pulls Far Ahead in the Race for AI Talent
“Will the United States fall behind China?” Jensen Huang, founder and CEO of US chip giant NVIDIA, recently stated in an AI-related program: “The answer is yes.”
Although the US has maintained dominance in the AI race over the past decade, boasting numerous top tech companies capable of developing cutting-edge models, China has caught up rapidly in recent years and is gradually showing signs of surpassing the US. This shift hinges on a critical factor in technological innovation: talent.
Chinese First Authors Surpass Those from the US and Europe
In December 2025, for the first time in history, the number of first authors from China in papers accepted by NeurIPS (Conference on Neural Information Processing Systems), one of the world’s premier AI conferences, exceeded that from the United States or Europe.
To better track international mobility trends among AI talent, The Economist analyzed the educational backgrounds of researchers who published papers at NeurIPS. The conference received more than 21,000 submissions, with roughly one quarter accepted. The magazine analyzed a random sample of 600 papers from 2025, authored by nearly 4,000 researchers.
In 2019, 29% of AI researchers publishing at NeurIPS began their careers in China. By 2025, this share had risen to 50%. Over the same period, the proportion of researchers starting out in the US fell from 20% to 12%.
By 2025, three‑fifths of Chinese AI researchers continued their advanced studies within China. Among the top ten universities where Chinese authors of 2025 conference papers earned their bachelor’s degrees, nine were based in China.
Graduates from Tsinghua University alone accounted for 4% of all researchers at NeurIPS; by comparison, graduates from the Massachusetts Institute of Technology (MIT), a top US institution, made up just 1%.
Open-Source Model Culture Drives Talent Motivation
The situation at NeurIPS encapsulates the growth of China’s scientific and technological talent pool.
Chinese researchers in the field may have stronger incentives to publish at top conferences, as China’s research ecosystem has a well-established reward structure for scientists, who typically need to list papers from leading conferences on their resumes.
Additionally, China’s culture of open-source models encourages authors to share their work in academic forums.
China’s appeal to international AI talent is also growing stronger. More and more AI professionals are choosing to build their careers in China.
According to data firm Digital Science, the number of active AI researchers in China now exceeds the combined total of the United States, the United Kingdom, and Europe.
Moreover, China’s AI talent pool is younger: 47% are students, compared with around 30% in Western countries.
China also places strong emphasis on STEM (science, technology, engineering, and mathematics) education: roughly two‑fifths of Chinese university students study STEM-related majors, nearly twice the share in the US.
Scale matters. Many Chinese graduates deliver cutting-edge innovative products, and a large cohort of AI-savvy researchers increases the likelihood of breakthroughs, meaning new technologies spread faster.
“China is building a highly skilled, well-trained workforce that is acutely attuned to AI development,” said Daniel Hook, head of Digital Science. “This points to a surge in AI companies emerging in China in the future.”
China’s Top Talent Could Double That of the US
Some of China’s finest innovations come entirely from homegrown talent.
For instance, DeepSeek, an innovative AI product developed in China and launched in January 2025, has impressed many competitors. All its core contributors hold no degrees from outside China.
These changes reflect China’s growing attractiveness to tech talent. Chinese universities are increasingly ranked among the world’s top institutions.
Meanwhile, China has launched numerous programs to attract outstanding researchers back from abroad, offering generous annual salaries, substantial research funding, and housing subsidies.
At the same time, the US has become less appealing to Chinese students.
Budget cuts and visa uncertainty have deterred prospective applicants, exacerbated by US doubts over the loyalty of tech talent. As a result, growing numbers are choosing to return to China.
To maintain leadership in AI, “winning the talent race is critical.”
If current trends continue, by 2028 China could have twice as many top-tier researchers as the United States.
(Translated by Wang Yue)
