In 2017, as China implemented strict new data localization measures that required companies to store Chinese data locally, within China, U.S.-based cloud companies like Amazon and Microsoft were forced to dramatically change their operations or, in some cases, even sell off parts of their Chinese business. It was a major sign that multinational companies might not be able to fully deliver on the promise of cloud computing, in which data is stored wherever in the world server space is cheapest and most readily available at any given moment, due to the ongoing tensions between China and the United States.
At the time, the Chinese data localization law was criticized by companies and regulators in the United States for restricting global information flows and, potentially, also making it easier for Chinese authorities to seize and search stored data relating to Chinese people and businesses, since they would have jurisdiction over all the data centers where that data was stored. It was also viewed as an unfair penalty for foreign cloud firms that did not already own infrastructure in China and would therefore have to spend significant sums of money to build data centers there if they wanted to continue to do business in China and compete with Chinese cloud giants like Alibaba and Huawei.
If one of China’s goals was to build up the strength of its domestic cloud providers, that strategy seems to be working. The New York Times reported this summer that the Biden administration was increasingly concerned about the dominance of Alibaba and Huawei in the cloud computing market, not just in China but also in other Asian countries and Latin America. These concerns were fueled by worries that the Chinese firms’ data centers might be used for espionage. Accordingly, the U.S. government began to make plans for trying to restrict the use of those companies’ technologies in the United States, and also lobbied other countries to not rely too heavily on Chinese providers.
Even as they worried about the growing market power of Chinese cloud platforms over the summer, the Biden administration also weighed possible rules restricting the ability of U.S. cloud platforms to provide Chinese companies with access to cloud services. Those rules would require U.S.-based cloud providers to get permission from the government before offering services to Chinese customers that made use of advanced artificial intelligence semiconductors, as part of the administration’s broader push to limit Chinese access to newer semiconductor technologies. This proposal is spurred in part by the perception that allowing Chinese companies to use U.S. cloud services has served as a means of enabling those companies to circumvent trade restrictions on sophisticated semiconductor technology.
The back and forth between U.S. and China on cloud computing platform rules and regulations highlights what a central roles cloud services play around many of the largest tech-related tensions between the two countries, including cyberespionage, artificial intelligence, and the provision of computing infrastructure for developing countries. These are issues that the two countries have wrestled with for several years, following many high-profile cyberespionage incidents that have heightened mistrust between U.S. and Chinese officials and spurred ongoing efforts by the United States to block Chinese companies like Huawei from providing hardware for new 5G wireless networks. Those efforts have also included U.S. officials lobbying other countries to block Huawei infrastructure, signaling the United States’ concerns about Chinese tech companies becoming popular technology vendors around the world due to the low prices of much of their hardware and software.
But unlike the efforts to ban Huawei 5G hardware, which are predicated in large part on the idea that removing physical hardware from mobile networks is a slow and expensive process that precludes installing equipment that might present any future security risks, U.S. efforts to restrict Chinese cloud computing services center on the notion that even the software services provided by these companies are too risky to be permitted. That is also the idea at the heart of proposals to ban Chinese-owned apps like TikTok and WhatsApp, which are often justified for similar reasons that focus on the potential for the data collected by these firms to be used by the Chinese government to spy on U.S. individuals and companies.
Because cloud computing has become so central to so many businesses, it is perhaps not surprising that it has become enmeshed in all of these controversies around cyberespionage, artificial intelligence, and tech sector competition. But it’s hard not to feel that this entanglement has dampened many of the potential advantages that the cloud computing model was supposed to be able to offer, particularly around flexibility, efficiency, and security. One of the primary benefits of cloud computing was supposed to be that it could seamlessly and dynamically move data around the globe, maximizing the efficiency of computing resources so that servers that usually lay dormant when one time zone went to sleep could be used at those times by people in other regions, rather than companies having to build data centers everywhere in the world. Another benefit was supposed to be that companies could rely on massive cloud providers, that had huge, dedicated security teams, to protect their data and computing networks for them.
To some extent, those promises still hold true, but they turn out to be much more measured than previously anticipated because of the extent to which countries like the United States and China have tried to restrict the global reach of both their own and each other’s’ domestic cloud providers. The remarkable success of cloud computing in so many areas can even be viewed as having helped cause these restrictions by placing it at the heart of many of the most controversial and critical technology policy debates between the U.S. and China.