The world is leaning on China a lot these days as a counterweight to the lunacy of Donald Trump. No one has illusions that China and its leader, Xi Jinping, are champions of democracy, but at a time when the US president is gleefully bombing boats and countries, and debating which regime to change next, China is an island of sanity.
I’ve made the analogy to Stalin in World War II, which continues to be appropriate. Roosevelt and Churchill had no illusions about Stalin’s USSR as a beacon of democracy, but they understood the essential role it played in defeating Hitler. China can play a similar role in protecting the world from the craziness emanating from the White House.
AI is one area where its role may prove to be extremely important. There have been many hugely overblown stories about how AI is going to take all the jobs and leave the rest of us unemployed and destitute.
This is an old theme about technology. Those of us who lived through the tech boom in the 1990s recall similar stories back then. There was even a boom in stories of technology-driven mass unemployment in the 1950s and 1960s. A famous novel of the time envisioned such a world in the not distant future. The fear that a new technology, in this case AI, will take all the jobs is not a new one.
Even if the prospect of mass unemployment is unlikely, there is a real concern that it will lead to even greater levels of inequality. Just to be clear, it is not the technology that creates inequality; it is the laws that govern its use. It’s unlikely that people would be making big fortunes on AI if the government didn’t grant patent and copyright monopolies to its developers.
But let’s leave that issue aside for a moment. The story of mass inequality is one where the AI makers are selling a product of enormous value that displaces millions of workers, including relatively highly paid workers. As a result, they can command huge profits from their AI.
Clearly, there is some validity to this story in that AI can displace labor in many areas, some of it highly paid. For example, AI can do much of the work in preparing legal briefs that is now done by lawyers. It’s not clear that AI will, on net, reduce the demand for lawyers, but it can substantially increase the productivity of lawyers.
But the fact that AI can lead to large gains in productivity doesn’t necessarily make the AI companies rich. That depends on the extent to which competition brings the price down.
To take an earlier technology, Dell is the largest manufacturer of computers in the United States. It is a successful and profitable company. Its market capitalization is less than $140 billion. That’s a good chunk of money, but less than 1/30th of Nvidia’s $4.8 trillion market capitalization.
The fact that the PC is an incredibly useful product that has hugely increased productivity has not meant that PC makers would get immensely rich and dominate the economy. The reason is that competition, even with weak antitrust enforcement, has forced down the price so that most of the benefits have largely gone to consumers.
This is where the Chinese AI makers come in. While the leading US makers may still be somewhat ahead by many measures, the Chinese companies are able to make AI products available to users, which likely meet most of their needs, at prices that are a fifth, a tenth, or even less than the price charged by the leading US companies.
For this reason, Chinese AI is beating out US in adoption through much of the world. Apparently, Chinese AI is even gaining many customers in Silicon Valley, both because of its lower price, but also because it is open source, which mean companies can alter it to fit their needs. This also means that a company can run the Chinese AI on their own systems and they don’t have to turn over control of sensitive company data.
This Chinese competition is a huge deal not only for bringing AI prices down, but also for preventing fascist clowns like Elon Musk from getting endless money. While Musk may always be insanely rich, if investors ever learn arithmetic and value his companies based on their profits, he will have far less money. (Tesla has a price-to-earnings ratio of 360. If it had a more normal, but still high PE of 20, Musk’s stake would be worth a bit more than 1/20th its current value.)
We should have that conversation about intellectual property rules that make the Musks of the world ridiculously rich. We should also be changing rules on things like bankruptcy that private equity barons to get rich by buying companies and putting them into bankruptcy.
Unfortunately, we have not yet advanced to the point where we can have a serious discussion on the ways we structure capitalism to generate inequality. Perhaps one day we will, but until then, we should be thankful for Chinese competition.
This first appeared on Dean Baker’s Beat the Press blog.
