Eric’s Review of AI Super-powers
Kai-Fu Lee recently published AI Super-powers: China, Silicon Valley and the New World Order, a fascinating inside perspective on the booming AI industry. The book lays out a vision of what problems AI might cause and what we as a global civilization can do to fix it. Lee presents a global future that is heavily dominated by the United States and China because he believes these two countries are dominating the AI industry. However, he notices many differences in style between the AI industries in both countries. According to Lee, the United States has a more idealistic style which is heavily focused on concepts of making the world a better place; in contrast, China has a more concrete and profit oriented style. Although Lee acknowledges that AI was essentially born in the West, he claims that China is making more innovation happen when it comes to implementing AI into real life situations. China is also excelling at applying AI to situations involving physical geography. Lee’s detailed account of China’s innovative applications of AI to the real world is very helpful for gaining an understanding of how to prepare for an AI future.
Lee also explains which occupations he expects to be automated soon. Those positions involving creativity and social skills are expected to remain, while impersonal and repetitive work involving maximizations based on data are expected to be automated. Lee takes note that social worker positions are currently undervalued and underpaid; this is one situation which he hopes will change in a new AI economy.
Lee expects a wide economic and power gap to emerge between the two AI superpowers and the rest of the world. He also expects the gap between the rich and the poor to expand dramatically. This wealth gap will expand because companies will increasingly hire less workers, fire others and pay many of the remaining workers less due to automated services. At the same time, AI will produce incredible economic outputs while leaving the profits
overwhelmingly in the hands of wealthy business owners. This problem can be seen as an extension of the phenomenon of outsourcing. Automation is even better than outsourcing for the profits of those who own the company.
Lee suggests several ideas to help humanity better deal with the burdens and difficulties of the coming AI age. He expects that universal basic income (giving all citizens a free basic income) will be the common solution that will be respected. However, Lee points out that work has more than just economic value. It also has a profound impact on the way we interpret the value of our lives. Work helps us to feel like valuable members of a community. It helps us to be socially connected and to have a healthy self-esteem. This is why Lee thinks UBI is inadequate. I’m inclined to agree that something more needs to happen on a policy level to address the problems that are coming. However, I am skeptical that Lee’s vision of the future will happen as he thinks it will.
One solution that Lee presents is that the government should pay social workers to do their jobs. The ability of humans to love, understand and comfort one another is unique to humanity; Lee emphasizes that this is the greatest thing about our humanity and that it makes us distinct from AI. I happen to agree with this position on humanity, but I also want a more concrete discussion about how to deal with the economic problems that will come with the AI revolution. Lee wants to pay these social workers through high taxes on the AI companies which he believes will generate huge profits of an unparalleled size. This sounds like a possibility for societies that already feel comfortable paying high taxes; but will low tax societies like Singapore, the US, Hong Kong or Mexico want to pay for this to happen?
Lee is from Taiwan, was educated to the Phd level in the United States, and has spent much of his AI career in China. My main criticism of the book is that it is far too confident that the future of the world will be determined by the parts of the world that Lee happens to be
familiar with, China, the US and the world of AI. I don’t believe that the world will be nearly as dominated by the US and China as he thinks it will be. I believe that the world’s future power structure will be much more multipolar and that Europe, South Korea and Japan will play significant roles as technological powers. A recent MIT Technology Review article titled Nine charts that really bring home just how fast AI is growing, showed more research papers on AI being published from Europe than from the US or China. This seems to go against Lee’s narrative that the US and China have a near monopoly over AI. MIT Tech Review is presenting a different picture of a tripolar AI world map, with Europe, the US and China as nearly equal powers in the AI revolution. However, despite Europe’s global lead in AI papers, Europe is spending less money on AI compared to the US and China.
In my opinion, other technologies besides AI might end up having comparable levels of influence on the future of humanity. Advances in particle physics, for example, may prove itself useful for huge new innovations. The energy industry is another field of technology with a huge influence on other technologies because AI and IT is ultimately dependent on the energy industry 100 percent for its own machines to function. If Europe, South Korea or Japan revolutionize the energy industry, I think that could potentially give them a lead over China and the US even if these two countries end up with superior AI. Changing the energy industry is the key to solving climate change, perhaps the world’s biggest problem.
Let me explain why I think the future of the world won’t be as Chinese and American dominated as Lee believes it will be. I think the world will be very multipolar, revolving around many different nations and technologies which will all be forced to collaborate with one another. I also think that China and the US have many social and political problems which will threaten their ability to dominate the world regardless of what happens to their respective tech industries. China has a serious problem with prohibiting freedom of speech and expression; the country has a poor ability to enforce the rule of law. China’s suppression of disagreement and frequent
use of state violence is unattractive to the kind of international support, financially or politically, that is necessary for a country to thrive in the 21st century world of globalization. Likewise the United States struggles with incredible amounts of social and economic inequality, to such an extent that the lower classes are not receiving the support they need for their young people to develop their skills. The United States also has an exceptionally high amount of debt relative to the size of its economy; and this debt is not just to domestic citizens. Much of the US debt is owed to foreigners, which means that it could drain American financial power and strength away from its own economy.
Europe has many great universities and research institutions, India has rapid economic growth, these facts present real challenges to the narrative of a future Chinese and American world with two superpowers dominating everything that happens on the global scene. The rise of Asia is not just about China. And Europe is hardly a disintegrating pile of ancient powers, it includes many top countries for technological innovation. Germany in particular is a country which has often been ranked first in the world for technological innovation and creativity. The EU’s higher life expectancy compared with the US may give it global credibility for immigrants with great brains looking for opportunities abroad. None of these criticisms diminish my respect for Lee’s impressive ability to explain the basics of AI to lay people. For this and other reasons, the book is well worth reading.
Links
https://www.technologyreview.com/s/612582/data-that-illuminates-the-ai-boom/ https://www.bloomberg.com/news/articles/2019-01-22/germany-nearly-catches-korea-as-innovat ion-champ-u-s-rebounds
https://aiindex.org/
https://tradingeconomics.com/country-list/government-debt-to-gdp
Book
Kai-Fu, Lee. AI super-powers: China, Silicon Valley, and the New World Order. New York New York. Houghton Harcourt Publishing Company. 2018. Print.