Rapid Ascent, Heavy Toll

Artificial intelligence remains at the heart of the technological rivalry between the United States and China. Chinese companies like DeepSeek are challenging American dominance. But as the country’s expertise grows, it has also faced a troubling wave of losses—five key AI figures have died within just three years. Officially, the causes include illness and accidents. Unofficially, questions linger about burnout, pressure, and the overall transparency of the system.

According to Chinese computer scientist Liu Shaoshang, AI professionals work under immense pressure despite high salaries. "The industry is moving too fast, and the competition is extremely fierce," he explains. Researchers face not only a race to publish, but also ethical dilemmas: the societal impact of AI remains poorly understood, intensifying moral stress.

As reported by the South China Morning Post, all five deceased specialists were at the peak of their careers and represented cutting-edge fields—from computer vision to military simulation and medical technology. Their deaths have left a noticeable gap.

In 2022, Sun Jian, Chief Scientist at Megvii (Face++), passed away at the age of 45. He had previously worked at Microsoft and was an expert in computational photography and computer vision systems. At Megvii, he led the development of the ShuffleNet neural network and the Brain++ platform, which helped the company enter the global AI arena. He also served as head of the AI faculty at Xi’an Jiaotong University. Holder of 35 U.S. patents, Sun was considered a pillar of the industry.

A year later, Feng Yanghe, a military simulation specialist and associate professor at the National University of Defense Technology, died. According to a military obituary, he passed away "on the way to an important mission." The War Skull I and II systems developed under his leadership became a foundation for strategic planning within the PLA. He had studied at Harvard and the University of Iowa, combining expertise in statistics and high-performance computing. He was 38 years old.

In the same year, 2023, Tang Xiao'ou passed away. He was the founder of SenseTime—one of China's "four AI dragons." Aged 55, he held a degree from MIT and had worked at Microsoft Research. His legacy included the advancement of computer vision in both civilian and commercial applications. SenseTime became known for its surveillance systems and generative AI. After Tang's death, the company’s stock plummeted, highlighting his irreplaceable role.

In 2024, He Zhi, co-founder and Chief Innovation Officer of Yidu Tech, passed away due to altitude sickness in Qinghai. He was 41. He oversaw the integration of AI into medical and insurance systems, merging biomedical sciences with big data. Previously, he worked at Alibaba and graduated from Tsinghua University.

In 2025, Kuan Yuhui, an associate professor at South China University of Technology and an expert in computational photography and texture analysis, passed away at the age of 39. He authored dozens of papers, was ranked among the top 2% of scientists globally by Stanford, and mentored a new generation of researchers. He studied and taught at SCUT and completed his postdoc in Singapore.

All of these deaths were officially attributed to illness or accidents. In the cases of Sun Jian and Tang Xiao'ou, it was illness—no diagnosis disclosed. He Zhi died from altitude sickness. Kuan's cause was reported simply as illness. Feng Yanghe was said to have died "in the line of duty." No signs of foul play were reported. However, in Feng’s case, the authorities’ silence raised suspicions: his death was announced with a ten-day delay using the euphemism "sacrificed his life." Chinese forums were flooded with questions about what really happened that night in Beijing, and whether the tragedy was tied to military infighting. No confirmation ever followed.

Tang’s death did not spark conspiracy theories—just shock at the loss of an exceptional scientist. His contribution to SenseTime was immeasurable. The deaths of Sun Jian and He Zhi were seen as tragedies amid immense work pressure. Kuan’s passing was mourned as the loss of an entire academic discipline and a generation of students.

What unites these stories is not only the age and stature of the individuals, but also the fact that all five were returnees from abroad—members of the "sea turtle" generation building bridges between China and the West.

These losses have left a painful mark on China’s AI sector. After Sun Jian’s death, Megvii lost much of its research momentum, and young scientists lost a mentor. Tang’s passing left SenseTime without a public face or scientific strategist, prompting a strategic reset. Feng’s loss disrupted military AI projects—it won’t be easy to find someone of his caliber. Yidu Tech may now lose pace in health tech. And academia lost a teacher who was shaping the future elite.

The problem goes beyond individual organizations. China is pursuing an ambitious AI strategy, investing billions into research and talent. The rise of DeepSeek—overtaking ChatGPT in the App Store—is a testament to this momentum. But if overload, stress, and burnout become embedded in the profession, they may deter the next generation. These losses risk not only slowing current projects, but also shaking confidence in the resilience of the entire system.

The American environment still appears more balanced: no comparable wave of deaths among AI leaders has been reported in the U.S. If China is to maintain its leadership ambitions, it must invest not only in technology but in people. Without sustainable working conditions, even the largest budgets won't prevent brain drain, lost momentum, or declining morale. The tech race isn't just about algorithms—it's also about those who build them.