In the hushed corridors of the technology hubs in Beijing, Hangzhou and Shenzhen, a quiet but significant shift is reshaping the landscape of AI development in China. Over the past 12 months, a wave of elite engineering and scientific talent has returned from the US to Chinese shores.
Led by the return of top researchers now leading AI development at tech companies ByteDance and Tencent, this reverse migration is raising a critical question: why are the architects of the future — most of whom stayed to work in the US after their studies — leaving Silicon Valley, the tech capital of the world?
For decades, the Valley served as a gravitational well for global tech ambition, a place where ideas met capital and engineering prowess. But the trajectory is starting to invert.
Figures such as Wu Yonghui, who gave up a senior role at Google DeepMind to lead ByteDance’s push into next-generation large language models, and Yao Shunyu, who left OpenAI to anchor Tencent’s AI development, have returned in the past year.
Other returnees include Roger Jiang, a senior scientist who left OpenAI to found his own robotics start-up in Shenzhen, and Zhou Hao, a researcher Alibaba poached from Google DeepMind to refine models. And three AI-focused headhunters based in China and San Francisco say they helped hire and relocate more than 30 US-based researchers to China in the past 12 months, versus a low single-digit a year earlier.
This is not just a trickle of nostalgia — it is a calculated realignment. Macro and micro factors in China are creating a magnetic pull that is becoming more difficult to resist.
High-growth companies Asia-Pacific
This feature is part of the High-growth companies Asia-Pacific Special Report.
Other articles include:
Start-ups evolve to reshape industries with AI
Indian winemakers braced for change
Comment: Trump is not the biggest threat to China’s global champions
On a macro level, the opportunity set in China is undergoing a transformation. While Silicon Valley debates the ethical dilemmas of AI applications, China is deploying AI across all sectors of its economy.
Its supply chain advantage is no longer just about assembling iPhones for Apple, but producing world-leading electric vehicles and robotic hardware, and creating vast amounts of quality data from its hyperconnected society.
From autonomous taxis in Beijing to AI-powered trading in Shanghai, AI implementation has soared. For a computer scientist, this represents a real-world laboratory where algorithms make large-scale impact almost instantaneously. While such implementations also happen in the US, regulations and security concerns are slowing their adoption.
“If you are in anything related to hardware, especially robotics, you have to be literally in Shenzhen,” says Steve Hsu, a professor of computational mathematics at Michigan State University who recently visited the city, which has at least 100 humanoid robotics companies. “If not, a piece breaks and you have to send it to Shenzhen and the turnaround is just too much.”
This macro opportunity is supported by more attractive micro conditions. Pay for top-tier AI researchers in China has surpassed Silicon Valley standards when adjusted for tax and cost of living, according to the headhunters.
They add that the purchasing power of a junior or mid-level researcher’s salary in a Chinese tech hub affords a lifestyle — including property, domestic help and access to world-class amenities — that feels out of reach for even well-paid engineers in the San Francisco Bay Area.
Efficient infrastructure, low crime rates and a sense of cultural fluency also create a gravitational pull that grows with age and family considerations.
Another recent returnee, Jonathan Zhou, a Harvard-graduated quant fund manager and father of a two-year-old, says: “I moved back to Shanghai because I think it’s a better place to build a family: China offers a more rigorous and merit-based education system which I prefer for my kid, and it is a safer place to live.”
However, the most decisive factor may be the “push” from the US. Rising geopolitical tensions and a more restrictive immigration regime have made shifting from student visa to green card an uncertain odyssey.
For the Chinese engineers who make up a significant portion of Silicon Valley’s AI labs and tech industry, it is clear the US no longer welcomes them.
Hsu, who gave lectures last month at China’s elite Tsinghua University, says: “You can also see an increasing amount of China’s brightest kids choosing to stay home, instead of pursuing [a] PhD in the US,” adding “it absolutely has something to do with the US immigration policies, in addition to just significantly more opportunities in China these days”.
Register for the 2027 ranking
The FT/Statista list of Asia-Pacific’s high-growth companies is published every April. Find out about the list and preregister for potential inclusion.
Lu Zhang, founding partner at Fusion Fund based in Silicon Valley, says it is too early to suggest the Valley is losing out to its Chinese competitors. The US, and specifically the Bay Area, still possesses the strongest ecosystem for technological incubation and development.
“The circulation of capital is the most efficient. Entrepreneurs get to validate their ideas quickly, and work with an extensive network of top-tier peers and mentors. You can’t find this anywhere else in the world,” says Zhang.
The recent poaching of Alibaba’s AI engineers by Meta is a testament to how the magnetic pull works both ways. For every researcher who returns to China, another may be lured to the Valley.
20%
Share of Tsinghua University’s engineering graduates who apply for US PhDs
A career officer at Tsinghua University, who declined to be named, says about 20 per cent of its engineering graduates still apply for PhDs in the US, down from about 50 per cent before Covid-19.
What does the return of the engineers signify? It shows the maturing of China’s tech sector. It no longer merely consumes US innovation; it creates it. For the returnees, their decision is not about rejecting Silicon Valley, but about embracing a new frontier.
They are betting that the future of AI will be written not just in the code that runs on servers in California, but in the data generated by billions of devices, factories and cities across China.
This migration is, at its core, a normalisation of the global tech order. Talent will flow to where the greatest opportunities and stability are perceived. For decades, that was a one-way street to the west. Now, with China accelerating tech development and the US tightening its borders, the traffic is beginning to flow both ways.
The winners in the global AI race will be those who understand that in the war for talent.