To The Moon

CFP

AsianFin -- The competition between the U.S. and China in artificial intelligence (AI) is taking a fascinating turn. What was once seen as a rivalry between two tech superpowers has increasingly become a contest between Chinese talent working in the U.S. and those in China.  

When Elon Musk’s xAI announced its Grok-3 model, two of the key scientists standing at the center of the spotlight were of Chinese descent. One was Wu Yuhuai, who hails from Jiande, Zhejiang Province, while the other, Jimmy Ba, remains somewhat of a mystery but is also of Chinese origin. In fact, out of xAI’s 12 founding members, five are of Chinese descent, including Greg Yang, a Harvard-educated mathematician from Hunan province, Guodong Zhang, a alumnus of(Zhejiang University and the University of Toronto, and Zihang Dai, a graduate ofTsinghua University and Carnegie MellonUniversity.  

This phenomenon has led some to joke that the so-called U.S.-China AI race should be rebranded as the “985 University Cup” AI Model Competition (2025 Season). 985 Universities is a collective term for China’s top universities.

The global AI talent pool has undergone significant changes in recent years. As shown in the Paulson Institute’s Global AI Talent Tracker 2.0, a growing number of the world’s leading AI researchers are now working in China, with Chinese researchers and institutions playing an increasingly prominent role in the global AI advancements. Chinese researchers now outnumber their American counterparts in the field, and nearly half of the world’s top AI researchers received their undergraduate education in China, underscoring the country’s education system as a critical talent pipeline.

In 2019, the U.S. dominated the AI talent scene, with 65% of top researchers working there, while China barely registered. Fast forward to 2022, and 12% of top AI researchers are now based in China, with 26% of them having completed their undergraduate studies in Chinese universities—second only to the U.S. (28%).  

Even within the U.S., Chinese researchers are making their mark. In 2022, Chinese nationals accounted for 38% of researchers in the top 20% of U.S. AI institutions, surpassing their American counterparts (37%). Globally, 47% of top AI researchers received their undergraduate education in China, compared to just 18% from the U.S. and 5% from India.  

In 2019, 18 of the top 25 AI research institutions were based in the U.S., including 14 universities and four tech companies—Google, Microsoft, Facebook (now Meta), and IBM. At the time, China had only two universities, Tsinghua and Peking, on the list, and both ranked in the middle.

By 2022, the U.S. had three fewer institutions on the list, with IBM dropping out. Meanwhile, China’s presence expanded significantly: Tsinghua and Peking University ranked among the top five, second only to Stanford and ahead of Carnegie Mellon and MIT. Additionally, institutions such as the Chinese Academy of Sciences, Shanghai Jiao Tong University, and Zhejiang University joined the rankings. Notably, Huawei became the first Chinese company to be included.

This shift highlights an important trend: AI research is increasingly driven by industry rather than academia. Universities alone cannot sustain AI research due to its enormous costs. While institutions like Tsinghua and Peking University may have budgets in the range of tens or hundreds of billions of yuan, these amounts pale in comparison to the financial resources required for large-scale AI development.

For instance, training GPT-4 reportedly cost OpenAI hundreds of millions of dollars, with electricity expenses alone reaching approximately $300 million annually. Even the best-funded Chinese universities cannot afford such costs.

This is why corporate investment in AI has become a game-changer. Companies such as Huawei, Alibaba, Tencent, ByteDance, and BYD have collectively poured billions into AI research, enabling the development of large-scale computing clusters and AI models. Without this corporate investment, China’s AI talent pipeline would struggle to sustain itself.

For example, Huawei has invested over $100 billion in R&D over the past decade, with a significant portion dedicated to AI. Alibaba recently announced plans to invest $52 billion in cloud and AI infrastructure over the next three years. Such investments have created an ecosystem where academic research and industrial innovation thrive together.

A few years ago, critics often questioned China’s education system, arguing that it stifled creativity and innovation. However, recent developments in AI research have disproven this notion. China’s AI boom demonstrates that when the country’s rigorous education system is combined with significant industry investment, it can produce world-class talent and groundbreaking AI advancements.

Whether in Musk’s xAI, OpenAI’s Sora team, or Chinese tech giants, Chinese-educated researchers are making their mark globally. Regardless of where they work, these AI pioneers received their foundational training in China’s education system.

As AI technology continues to evolve, one thing is clear: the so-called China-U.S. AI competition has, in many ways, evolved into a contest between Chinese AI talent at home and abroad. With China’s education system and industry working in tandem, the country is poised to remain a global AI powerhouse for years to come.