AI Superpowers

In 1999, Lee gave a lecture on speech and image recognition at the University of Science and Technology of China. He recalls that, while the computer science students were attentive and hardworking, they were hindered by limited access to tech and high-quality learning materials. In the span of just 30 years, China evolved from this to an AI superpower. Lee writes:

[C]reating an AI superpower for the twenty-first century requires four main building blocks: abundant data, tenacious entrepreneurs, well-trained AI scientists, and a supportive policy environment […] This chapter assesses the balance of power in the two remaining ingredients—AI expertise and government support (92).

Throughout this chapter, he argues that China has all of these elements in greater abundance than any other country. He believes that, up to this point, genius “elite researchers” were the key to leading the world in AI. Now, the economy will favor larger workforces of “solid AI engineers.” While the US has the former, China’s high population and cultural infatuation with AI ensures the latter.

Lee also notes that, where America has a “combative political system [that] aggressively punishes missteps or waste in funding technological upgrades, China’s techno-utilitarian approach rewards proactive investment and adoption” (93). While the Chinese government sometimes restricts innovation in the social sciences, Lee writes that its control over all aspects of digital life within its borders incentivizes it to nurture AI research.