Enjoyed the episode? Want to listen later? Subscribe here, or anywhere you get podcasts:

There’s this assumption that because you have an AI plan, you’ve solved the problem. Whereas actually central forecasting, and planning about complicated, uncertain technological developments is oftentimes flawed and doesn’t solve anything. It may actually be counterproductive.

Jeff Ding

The State Council of China’s 2017 AI plan was the starting point of China’s AI planning; China’s approach to AI is defined by its top-down and monolithic nature; China is winning the AI arms race; and there is little to no discussion of issues of AI ethics and safety in China. How many of these ideas have you heard?

In his paper ‘Deciphering China’s AI Dream’ today’s guest, PhD student Jeff Ding, outlines why he believes none of these claims are true.

He first places China’s new AI strategy in the context of its past science and technology plans, as well as other countries’ AI plans. What is China actually doing in the space of AI development?

Jeff emphasises that China’s AI strategy did not appear out of nowhere with the 2017 state council AI development plan, which attracted a lot of overseas attention. Rather that was just another step forward in a long trajectory of increasing focus on science and technology. It’s connected with a plan to develop an ‘Internet of Things’, and linked to a history of strategic planning for technology in areas like aerospace and biotechnology.

And it was not just the central government that was moving in this space; companies were already pushing forward in AI development, and local level governments already had their own AI plans. You could argue that the central government was following their lead in AI more than the reverse.

What are the different levers that China is pulling to try to spur AI development?

Here, Jeff wanted to challenge the myth that China’s AI development plan is based on a monolithic central plan requiring people to develop AI. In fact, bureaucratic agencies, companies, academic labs, and local governments each set up their own strategies, which sometimes conflict with the central government.

Are China’s AI capabilities especially impressive? In the paper Jeff develops a new index to measure and compare the US and China’s progress in AI.

Jeff’s AI Potential Index — which incorporates trends and capabilities in data, hardware, research and talent, and the commercial AI ecosystem — indicates China’s AI capabilities are about half those of America. His measure, though imperfect, dispels the notion that China’s AI capabilities have surpassed the US or make it the world’s leading AI power.

Following that 2017 plan, a lot of Western observers thought that to have a good national AI strategy we’d need to figure out how to play catch-up with China. Yet Chinese strategic thinkers and writers at the time actually thought that they were behind — because the Obama administration had issued a series of three white papers in 2016.

Finally, Jeff turns to the potential consequences of China’s AI dream for issues of national security, economic development, AI safety and social governance.

He claims that, despite the widespread belief to the contrary, substantive discussions about AI safety and ethics are indeed emerging in China. For instance, a new book from Tencent’s Research Institute is proactive in calling for stronger awareness of AI safety issues.

In today’s episode, Rob and Jeff go through this widely-discussed report, and also cover:

  • The best analogies for thinking about the growing influence of AI
  • How do prominent Chinese figures think about AI?
  • Cultural cliches in the West and China
  • Coordination with China on AI
  • Private companies vs. government research
  • How are things are going to play out with ‘compute’?
  • China’s social credit system
  • The relationship between China and other countries beyond AI
  • Suggestions for people who want to become professional China specialists
  • And more.

Get this episode by subscribing to our podcast on the world’s most pressing problems and how to solve them: type 80,000 Hours into your podcasting app. Or read the transcript below.

Producer: Keiran Harris.
Audio mastering: Ben Cordell.
Transcriptions: Zakee Ulhaq.

Highlights

Why Cold War analogies don't work with AI

So I think there will be competitive dynamics as it relates to specific AI enabled military applications, [but] I don’t think it will be on the same scale as the arms race in the Cold War. I think in the Cold War there were a bunch of unique dynamics, but one aspect that was unique is there was something that was countable about ICBMs, nuclear weapons, that made it so we were always thinking that there were some missile gap between the US and the Soviet Union. For AI, it’s not necessarily the case that we’re just counting how much AI is in the military. So in that sense it’s more about a system-wide transformation about how you upgrade your military across a whole range of factors from information, logistics, communications. So I think that’s a poor analogy. And then the other difference is that it’s not just about the weapons applications that will be competitive, it’s about these system-wide applications: who will be able to adopt AI across different manufacturing systems to enable wider growth in the entire economy.

China's social credit system

One thing that I’ve been reading about a lot lately is the social credit system. So this ties into a lot of the things that we’re talking about in which people who have researched what’s actually happening in the social credit system, for example, Jeremy Daum at Yale Law School, he says that when Western servers look at what’s happening in China and the social credit system, they’re looking through a glass darkly. That they’re projecting their own worst case fears about what the social credit system could become and then saying that’s what’s happening in China. Whereas actually other people who have been researching the social credit system say it’s a very low tech system of black lists, credit scoring that isn’t using machine learning algorithms for now, even though there are vague claims that they’re going to integrate AI and big data into these systems. So I think that is another misperception. The idea that the social credit system is a very high tech AI empowered system where maybe that is part of the vision or part of these vague proclamations, but it’s definitely not what’s happening on the ground right now.

Should you learn Mandarin?

My advice would be if you have some of the language already, or you have a strong passion to learn the language or you like learning languages, to invest in learning Mandarin as much as you can, especially if you want to learn about tech in China. Because for me, I read probably 60 to 40 of my stuff, in terms of learning about China’s AI development, is Mandarin to English. So I think some of the best stuff is still in English, but there’s just so much more stuff coming out every day in Mandarin about what is happening in China’s AI scene. It’s like if you’re a German, you’re not going to read Der Spiegel about what’s happening in terms of Silicon Valley. Even if Der Spiegel has a correspondent in Silicon Valley, sure, they may have like three people covering Silicon Valley. In the US, you’re going to read MIT tech review, you’re going to read Bloomberg news, you’re going to read The New York Times. So if you want to learn about what’s happening in China’s AI scene, sure those New York Times correspondents, Financial Times correspondents across China, they’re doing great work and they’re getting stories that you can’t get in some of China’s censored media. But for a massive scale effect, you should be reading Chinese language news or translations of Chinese language news.

[However], honestly, the neural machine translation is getting really, really good and it can augment translation capabilities to the point where you can build pipelines, and I think the Center for Security and Emerging Technology, one of the organizations I work with, is trying to build a more automated industrial scale pipeline for translation of these materials. I’m sure these exist across other institutions and I think it can get to the point where you can be someone who knows none of the language, but you have read all the best English language work on this and then you can say, “I want a translation of this thing” and then use that. Feed it through the pipeline and then work from there. Yeah, so I’m a little bit agnostic between the choices, but just know that there is a trade off.

Articles, books, and other media discussed in the show

Jeff’s work

Everything else

Related episodes

About the show

The 80,000 Hours Podcast features unusually in-depth conversations about the world's most pressing problems and how you can use your career to solve them. We invite guests pursuing a wide range of career paths — from academics and activists to entrepreneurs and policymakers — to analyse the case for and against working on different issues and which approaches are best for solving them.

The 80,000 Hours Podcast is produced and edited by Keiran Harris. Get in touch with feedback or guest suggestions by emailing [email protected].

What should I listen to first?

We've carefully selected 10 episodes we think it could make sense to listen to first, on a separate podcast feed:

Check out 'Effective Altruism: An Introduction'

Subscribe here, or anywhere you get podcasts:

If you're new, see the podcast homepage for ideas on where to start, or browse our full episode archive.