How American companies are fueling China’s AI race

Lenin is quoted as saying, “the capitalists will sell us the rope with which we will hang them.” Wittingly or unwittingly, American tech companies are selling rope to China in the form of an intellectual property transfer, the likes of which has never before been seen.

Companies like Microsoft, Apple and Amazon have contributed to China’s remarkably rapid development of artificial intelligence capabilities — all at the expense of American interests and security. Following the recent World Artificial Intelligence Conference in Shanghai, where Alibaba and Huawei showcased a new AI image generator and AI model upgrade, respectively, the repercussions have never been clearer.

The United States and China are in the midst of a high-stakes competition for AI dominance as the technology’s transformation of global political, economic and military power accelerates. The U.S. government has expressed deep concern about the Chinese government’s AI program, outlined the cyber threat posed by China and documented the Chinese Communist Party’s (CCP) human rights abuses. Even with tensions between the two countries rising, several American companies have doubled down on their AI investments in China.

For example, one recent report revealed that American technology firms and investors — including the investment arms of Intel and Qualcomm — poured $40.2 billion into Chinese AI companies between 2015 and 2021. Meanwhile, AWS continued hiring in China last year to build AI, even though Amazon’s CEO has been publicly arguing that “the U.S. [is] at risk of losing its technology leadership position to China as AI developments advance there.”

Microsoft, meanwhile, has partnered with researchers affiliated with China’s National University of Defense Technology on AI research that could be used for surveillance and censorship. Despite concerns that this technology would further the CCP’s oppressive tactics on its citizens, Microsoft said the research was “published to ensure transparency so that everyone can benefit from our work.”

Indeed, Microsoft Research Asia (MSRA) has been called “the seed capital” from which Chinese AI researchers and companies have developed — and that’s because thousands of MSRA-trained Chinese researchers have gone on to work at some of China’s major AI companies. These include Baidu, which is of course rolling out Ernie, as well as Alibaba and its own ChatGPT rival, Tongyi Qianwen.

This assistance to China is not just an issue for American tech companies; it also encompasses the development of competing tech talent more broadly. Though many Chinese students who study AI in the United States have stayed here to work after graduation — a benefit to our economy — the Chinese government has an estimated 200+ recruitment programs working to bring its homegrown talent and the AI knowledge these students have learned in the United States back to China.

Simply by way of example, just last month, the Department of Justice charged a former California-based Apple employee with stealing trade secrets from the company’s self-driving car division after leaving to work for the American subsidiary of a Chinese company. Following law enforcement’s search of his home, he purchased a one-way ticket to China and fled that night.

Perhaps American companies are beginning to realize their mistakes. Instead of continuing to “sell rope” to China, they are looking for a way to mitigate their exposure to the Chinese market. Apple is taking steps to de-risk its supply chain from China by increasing manufacturing in India and in the U.S., and Microsoft sources recently revealed that the company is now seeking to move its top China-based AI experts to its research institute in Vancouver.

These actions seem to be a quiet acknowledgment of the risks of partnering with China. But they may be too little, too late. Broader commitments and investments in China continue to leave many companies in a long-term bind with no clear path out of their dependence on the country.

American tech companies are chasing profits by working with an oppressive regime that has already used AI to advance state surveillance and conduct grave human rights abuses on an unprecedented scale. At the same time, they’re compromising their own intellectual property, long-term competitiveness and American leadership in a critical new field of technology. 

For many, the rope has already been sold and the noose has been fashioned — it remains to be seen if they can escape the hangman.

Paul Rosenzweig served as deputy assistant secretary for policy at the Department of Homeland Security. He is a professorial lecturer at The George Washington University Law School and senior fellow in the Tech, Law & Security Program at the American University, Washington College of Law. The founder of Red Branch Consulting PLLC, he is also senior adviser to The Chertoff Group.