The US Strategy To Constrain China In The AI Sector Will Further Intensify – Analysis

By Zhou Chao

Since the Trump administration, the U.S. has gradually increased its restrictions and blockades on China in the field of high-tech industries, with the aim to curb the development speed of China’s advanced technologies and maintain its own advantage over China. This approach has continued into the Biden administration without change. The same is true in the rapidly growing field of Artificial Intelligence (AI) in recent years. However, within the AI sector, the U.S. seemed to show a certain degree of willingness for cooperation with China at one point.

From 2023 to the first half of 2024, in exchanges at various levels between the two countries’ governments, the U.S. side consistently expressed a willingness to cooperate with China on the global AI regulatory framework. The most obvious example of this was the official invitation extended to the Chinese delegation to attend and speak at the first Global AI Summit held in November 2023 in the United Kingdom. Considering the close relationship between the U.S. and the UK, it would have been difficult to imagine the Chinese delegation’s attendance without U.S. permission. Subsequently, high-level government dialogues on AI were held between the U.S. and Chinese governments in Switzerland. For a moment, it seemed that China had returned to the global stage in the AI field. However, based on a comprehensive analysis of various factors, researchers at ANBOUND believe that the “tech Cold War” between the two will persist and intensify, with no substantial shift in sight, particularly in the AI field.

The significance of AI technology makes it inevitable that the U.S. will take every measure to suppress China’s advancements in this field. Since the advent of modern science and technology, one of humanity’s central challenges has been to harness technology in a comprehensive and efficient way to support both production and everyday life. Throughout the first three industrial revolutions, and even in the ongoing fourth technological revolution, new technologies have yet to fully free people from monotonous, repetitive labor to allow them to focus solely on creative endeavors. However, the rapid development of AI, particularly generative AI, has made this long-sought goal a tangible reality. Research highlights that through automation and smart technologies, AI can take over tedious, repetitive tasks, freeing up time and energy for more complex and creative work. AI’s potential to automate production processes, optimize supply chains, enhance product design, and more can further boost productivity, creating more space for human innovation.

Beyond its impact on productivity, AI also plays a critical role in advancing scientific research and technological innovation. It can assist scientists with large-scale data analysis and simulations, accelerating the pace of discovery. It can aid in the development of new materials, the creation of new pharmaceuticals, and the solution of complex scientific challenges, driving progress in various fields. In military applications, AI also holds great promise for enhancing efficiency and combat capabilities. Such transformative potential in reshaping human labor and daily life is truly disruptive. The country that gains a dominant position in this technology will inevitably rise as a global leader in both political and economic spheres. Given that China is already recognized as a systemic competitor, the U.S. cannot afford to allow China’s AI technology to fully flourish and will continue to aggressively suppress it.

Secondly, the development of China’s AI industry and technology in recent years indicates that the U.S. measures to counter it need to be further intensified. In the first half of 2024, a report released by MacroPolo, a think tank under the Paulson Institute, showed that while the U.S. remains the undisputed global leader in AI talent, China has made remarkable progress in terms of talent cultivation and the prospects for talent development.

In 2022, the proportion of top global AI researchers trained in China rose to 47%, a significant increase from 29% in 2019. AI talent from China has also made notable improvements in qualifications. Among the top 2% of elite AI talents, 26% come from China, while the U.S. accounts for 28%, with both countries being very close. In the previous survey in 2019, only 10% of the most elite AI talents were from China.

In terms of job choices, the previous survey showed that only 11% of top AI talents (the top 20%) chose to work in China, while the U.S. employed nearly 60% of the world’s top AI talents. However, the 2024 survey revealed that 28% of top AI talents have chosen to work in China, while 42% are working in the U.S. This indicates a significant narrowing of the gap between China and the U.S. in recent years, with the choice of workplace for AI talent shifting from U.S.-dominated to a competitive dynamic between the two countries.

Regarding research institutions, in the previous list of the top 25 AI research institutions compiled by MacroPolo, based on the affiliations of paper authors, Tsinghua University and Peking University were the only two Chinese institutions. But in the 2024 statistics, Chinese research institutions now hold 6 positions, with Shanghai Jiao Tong University, Zhejiang University, the Chinese Academy of Sciences, and Huawei also making it into the top 25. Tsinghua and Peking University ranked 3rd and 6th, respectively. Previously, among AI researchers who received their undergraduate education in China, about 56% chose to pursue graduate studies in the U.S. However, according to the 2024 report, this figure has dropped to around 36%. Since graduate education is closely linked to future career destinations, this trend suggests that more AI talent may choose to work in China in the future.

The report suggests that these trends are largely driven by the rapid growth of China’s AI industry, which has led to increased demand for talent, as well as the growing emphasis by Chinese universities and tech companies on AI talent cultivation and recruitment. As things stand, the increasing strength of China’s AI capabilities is the primary reason for the shifting trends in the flow of top AI talent. Although the U.S. still ranks at the top globally in terms of AI research strength and international appeal, China’s rising influence means that the U.S. will need to take act to contain the development of China’s industry and research while strengthening its own research capabilities and global attractiveness.

In the specific area of hardware, Huawei successfully launched a 7-nm chip phone last year. Although in-depth analysis suggests that the breakthrough was mainly due to the use of older-model photolithography machines for multiple exposure and etching operations, it remains a fact that China’s research and development capabilities continue to advance despite multiple rounds of sanctions. Therefore, the U.S. needs to intensify its efforts to more effectively curb China’s progress in AI technology.

Moreover, the U.S. business community’s ties to China have gradually loosened, making the resistance to further sanctions less significant. For a long time, the U.S. business sector has been a key opposing force to U.S. sanctions on China. Due to the substantial investments and profit interests that American companies have in China, the business community has consistently opposed deep and comprehensive sanctions against China.

Take Nvidia, a key player in the AI field, as an example. Its president, Jensen Huang, repeatedly emphasized in 2022 and 2023 that the Chinese market is crucial, even indispensable, and that Nvidia would not leave the Chinese market. After the U.S. escalated its sanctions, Nvidia continued to export to China, even if it meant cutting down the functionality of its chips. However, data from the 2023-2024 fiscal year revealed that although revenue from Mainland China accounted for 21% of Nvidia’s total annual revenue, it dropped to single digits in the fourth quarter of the 2023-2024 fiscal year, with the proportion likely to continue decreasing. Despite the decline in revenue share from Mainland China, it did not hinder Nvidia’s overall revenue growth. Therefore, the importance of the Chinese market for leading U.S. AI companies is gradually diminishing. Additionally, China’s recent calls for increased use of domestically produced chips will undoubtedly further erode the confidence of foreign AI companies in the Chinese market.

For the broader U.S. business community, the importance of the Chinese market is also steadily diminishing. In 2023, American and Western companies withdrew over a hundred billion dollars in profits from the Chinese market, and investment institutions have been continuously scaling back their investments. This also means that when the government is ready to implement deeper sanctions on China, the opposition pressure from the business community will be more limited.

According to a report from The New York Times in the first half of 2024, over 100 tech leaders from Silicon Valley were lobbying Congress at that time, urging lawmakers to impose more restrictions on China’s AI sector. Among them were Palantir CEO Alex Karp and Sequoia Capital managing partner Douglas Leone. Palantir, a leading U.S. data analytics and AI company, is also a major defense contractor for the U.S. government and is rumored to have assisted the U.S. military in locating and killing Al-Qaeda leader Osama bin Laden. The New York Times reported that this lobbying effort from Silicon Valley not only sought to push Congress to limit China’s progress in the AI field but also aimed to improve AI regulations, increase technology research and development funding, and ease immigration restrictions to attract more international AI talent to the U.S. These lobbying moves are undoubtedly targeted primarily at China, aiming to both technologically block China’s progress and to intensify the brain drain from China, thus hitting China’s AI technology development from both fronts.

ANBOUND has previously pointed out that China’s stringent regulations in the AI sector are detrimental to the development and application of technology. The overall level of openness and social freedoms in China also hampers further advancements in AI technology. The current state of AI technology development in China mainly benefits from the accumulation of previous years and remains in a relatively static state. However, as U.S. sanctions deepen and become stricter, particularly with targeted measures against China, the development of technology and hardware will be further constrained, and the trend of talent outflow will become more pronounced. As a result, China’s AI industry may face even greater difficulties in its development.

With the Democratic Party’s looming electoral defeat last year, the Biden administration began in the fourth quarter of 2024 to escalate its blockade and crackdown on China’s chip industry, with even mature chip manufacturing equipment now being included in the sanctions. As a critical support for AI technology, the continued tightening of chip industry sanctions is undoubtedly a clear indication of the further tightening of the U.S. noose around China.

Final analysis conclusion:

As a global technological trend, China and the U.S. will continue to cooperate in the field of AI regulation. Even after Donald Trump formally took office, the AI dialogue mechanism between China and the U.S. was expected to continue. However, this will not alter the growing trend of the U.S. tightening its AI noose around China. During the Cold War, the U.S. and the Soviet Union cooperated in preventing nuclear proliferation, yet the U.S. never ceased its efforts to suppress Soviet advancements in nuclear technology. In the context of the U.S.-China technological rivalry, some have described the U.S. approach as a “small yard, high fence” on current trends and future directions, the yard may not necessarily become smaller, but the fence will inevitably continue to grow higher.

• Zhou Chao is a Research Fellow for Geopolitical Strategy programme at ANBOUND, an independent think tank.