China is locking down AI talent. Why does this matter?
Bloomberg reported on May 26 that Beijing was imposing travel restrictions on AI professionals, including startup founders, researchers and executives, at private firms such as Alibaba and DeepSeek, extending controls long associated with sensitive researchers and state-owned enterprise executives.
Furthermore, Beijing added new outbound investment rules (translation), published June 1 and taking effect July 1, that restricts ‘dispatching technical personnel across borders, organizing personnel to work in other countries (or regions), providing technical guidance across borders, or arranging cross-border training of personnel’ in relation to prohibited exports.
In the context of the heightened geopolitical competition over AI, China is beginning to treat elite scientists and engineers as strategic assets, to be governed just as other strategic assets are.
The Challenge for Australia
For Western liberal democracies such as Australia, this matters enormously. Domestically, Canberra is looks to ‘capture the opportunity’ in AI by building smart infrastructure, backing domestic AI capability, and attracting global investment. It is looking, with the United Kingdom and the United States, to develop capabilities in AI as one of the components of AUKUS Pillar II.
Talent is, along with hardware and compute, critical to this endeavour. Not only talent at universities and government institutes, but talent that resides with private companies who are at the centre of ‘increasingly industry-driven’ frontier research.
The growing implication of this is that these restrictions matter. They matter not only because this means a reduction in the potential flow of talent and capital out of China into Australia and other countries in the region. It furthermore matters because of the fundamental asymmetry of handling AI talent between liberal, Western, democratic societies and societies like China, simultaneously an asset but also a weakness to be exploited.
Firstly, it is unthinkable that Western governments should impose restrictions to the same level as China has done. The regulation of exports, scrutinising of foreign investment, and control of classified information and scrutiny of the movements of those who hold security clearances are commonplace enough. But a private-sector AI researcher, even one working close to frontier capability, is ordinarily free to leave their employment if they find a better post.
That freedom is part of Western governmental and institutional strength. The United States, Britain, Canada, and Australia became scientific hubs because talented people could come, work, collaborate, and leave.
The answer for further development is not to copy Beijing by turning open societies into gilded cages.
Although the West appears currently to hold an advantage in AI talent—and there have not been, as far has been publicly reported, any high-profile movements from Western to Chinese AI companies—it cannot treat talent mobility as a non-problem.
This is because China understands pull factors. State-backed programs like the Young Thousand Talents offer greater funding and larger research teams. The Charles Lieber case is instructive as a straightforward case of attraction: sentenced for concealing a Chinese Thousand Talents Program affiliation; Lieber is now rebuilding his work in Shenzhen with resources he did not have at Harvard. Song-Chun Zhu, on the other hand, is a case of return: he moved from UCLA to China in 2020 because the latter could offer a ‘lavishly funded Beijing Institute for General Artificial Intelligence’ while in the US his ideas were consigned to the margin and his own identity became a source of suspicion.
What Can Be Done?
The Western response should be two-pronged: Australia would likely work within the AUKUS framework, perhaps with an AI-oriented equivalent to proposals for an AUKUS Visa that would seek to keep talent within the ecosystem while enabling locational flexibility.
First, the straightforward improvement of hard reasons for researchers to stay: greater public research funding and resource provision, maintaining the open culture of Western societies, and imposing (openly) few bureaucratic or national security hurdles to movement. ‘Professional opportunities…lifestyle and culture, the political climate, and personal relations’, according to a 2021 study, shape where top researchers want to work. Talent flourishes when given resources and colleagues: the West should seek to be the best location for both.
But, and perhaps more controversially, it would be increasingly important to treat ‘flight risk’ as real in strategically deciding on hiring, training and education opportunities. This does not mean the counterproductive discrimination by ethnicity or nationality (see the Song-Chun Zhu case mentioned earlier, who left the US amidst the widely-criticised China Initiative of the Biden administration). It means acknowledging that decisions about access to elite mentors, compute, proprietary datasets, and frontier models have downstream consequences. Top talent is not a mere selfless contributor: their environment also improves their own skills and knowledge. All these activities, moreover, are catalysed by Western liberal (and open) societies and capital markets which are uniquely placed to provide those advantages. Where the public pays to train or enable strategic talent, it may reasonably ask for transparency, cooling-off periods, repayment clauses, or post-employment reporting.
China’s restrictions may help it lock in key AI personnel. It is not too late for the West to come up with a coherent and systematic response of its own. Ambitious scientists should bear in mind that returning to China can mean losing the freedom to leave and accepting other risks. Earlier returnees such as nuclear scientists Yao Tongbin and Deng Jiaxian, who returned to China post-1949 and faced persecution in the Cultural Revolution of the 1960s-1970s, learned that being embraced by Chinese politics could carry terrible personal costs. Unless liberal societies take talent retention seriously, they will preserve the liberty of their best minds to depart while rivals reserve the power to keep theirs home.
Alexander Yen is a DPhil candidate in International Relations at the University of Oxford and Founding Convenor of the Oxford China Research Group. His research focuses on Chinese foreign policy, state rhetoric, international order, and strategic competition. He has written on China, AI, and geopolitics for policy-facing audiences.
This article is published under a Creative Commons license and may be republished with attribution.
