Podcast with Edward Parker, Physical Scientist at The RAND Corporation

My guest today is Edward Parker, a physical scientist at the RAND Corporation. Edward recently authored a report titled “An Assessment of the US and Chinese Industrial Basis in Quantum Technology” and we discuss its key findings, what surprised him the most during his research, and much more.

Listen to additional podcasts here

‍

THE FULL TRANSCRIPT IS BELOW

Yuval: Hello, Edward. And thanks for joining me today.

Edward: Hi, very happy to be here. Thank you.

Yuval: So who are you and what do you do?

Edward: My name is Edward Parker. I'm a physical scientist at the RAND Corporation. So very briefly, RAND Corporation is a non-profit, non-partisan public policy research organization, which does research on a very wide variety of topics. Everything from national security to healthcare, to justice policy, Homeland security. And we have a large number of researchers with a very diverse variety of backgrounds. And we pride ourselves on working on a large number of topics in the public interest and putting together very multidisciplinary teams to work on those projects. So I myself am a physicist by background. I got my PhD in hard condensed matter physics at UC Santa Barbara at the end of 2017 and have been working at RAND since then on a variety of issues, mostly relating to emerging technologies and their impact on national security, although not entirely. So I myself spend most of my time working on quantum technology in particular and public policy issues around that, although I've also researched a little bit on artificial intelligence and 5G as well.

Yuval: And I think recently you were the lead author on a report about quantum technologies comparing the US to China. Is that correct?

Edward: Yes, that's right. So in February I believe, we published a report called An Assessment of the US and Chinese Industrial Basis in Quantum Technology, which is available for free on our website at rand.org with a large multidisciplinary team that consisted of physicists and political scientists, engineers, graduate students, Chinese language experts, with the goal of putting together a set of repeatable, consistent, and objective metrics which could be applied to any nation to give a repeatable and as quantitative as possible sense of what does their industrial base look like overall. And then we did two in depth case studies on the United States and China. But this framework is repeatable and in principle could be applied to any country. And as part of that, we took a very multidisciplinary mixed methods approach, which is something that RAND takes pride in. And we looked at really every aspect of this ecosystem.

So not only the private industry, the sort of traditional industrial base one might think of, but also academia and government support for quantum technology and specific technology achievements that have been reported in the literature. And the reason we did that was because quantum technology is still very new. There wasn't really very much of a private industry in quantum to speak of more than five or 10 years ago. And so unlike a more mature sector where most of the R&D is done in private companies, there's still a very important part to play in academia, open scientific research, national labs, government support. So we didn't think we'd get a complete picture by just looking at private industry so we tried to cast our net as wide as possible.

Yuval: Before we jump into the conclusions, I'm still curious about the methodology. I would guess that the US is much more open in sharing information, in being able to access information than China. So how do you know that the data that you're using to assess the Chinese quantum industrial base is correct? That there's not a lot that's hidden that will pop up one day and say, "Oh, we have a computer that can use Shor's Algorithm to break encryption” tomorrow morning? How do you know that you're getting good data?

Edward: So you're absolutely right that data availability is a significant limitation when everyone looks at any country, including the US. We don't have perfect data on the us either. So overall, we had more difficulty accessing data on China, perhaps not surprisingly. I suspect the Chinese themselves also perhaps have less information on their own industry as they're sort of less of an organized data collecting organizations. There's no analog of the quantum economic development consortium in the US, for example. So the data availability issue varied depending on what we looked at. So one of the main methodologies we took was looking at open academic literature. Now, that was only in the English language literature so there's already a limitation there. But we think that we have a fairly comprehensive look at the open scientific literature because from our assessment and in consultation with our internal experts, most of the most groundbreaking research we believe is published in English language journals. Science, Nature, physical review letters, and so on which we had access to.

So we do think we have a reasonably representative snapshot of their open scientific publishing. It's harder to get data on the private sector activity, certainly. And of course, we don't know what they are choosing not to report, but even with the private sector companies, there was a surprising amount of information available online. There were financial documents reported. They were in Chinese, and we did have Chinese language expertise on our team, which helped that. But we think that we were able to get, I should say as reasonable of impression as one can get based on the open literature and information available online subject to the limitations of resources and time, of course. But my sense, hard to say for sure, is that with quantum, since most of the technology is fairly young and most of the technology is still being developed, there's a reasonably strong incentive, even by the part of companies to publish their progress, if nothing else than to encourage venture capital investors to invest more money in them.

And while there's certainly some trade secrets, of course, my guess is that you don't have something on the level of an entire quantum program that is completely not public. That is again, just a guess, but we have circumstantial evidence there. So I should also mention we actually deliberately put slightly more effort into, more focus into the US assessment than the Chinese assessment. That was a choice of research prioritization. So for a few of the metrics we do simply report we did not assess this on the Chinese side, or we did not have as comprehensive of an assessment as the US side,

Yuval: When one looks at quantum technologies, typically there's a division into three areas. And I think that's also reflected in your report. Quantum computing, quantum sensing, and quantum communications. So if we were scoring the quantum Olympics, where does the US come first? Where does China come first in each of these areas?

Edward: So it's a complicated question and I could go into more depth on each of the three. But broadly speaking, in quantum communications I think China is ahead both in terms of volume of high impact academic literature published as a whole, as well as specific technologies demonstrated.

For example, China is the only country that we know of who have demonstrated a quantum communication satellite capable of doing something called quantum key distribution from space. I'm happy to go into more detail if you'd like. They've also laid down a very, very widespread fiber optic network of quantum key distribution technology, which very briefly is the technology which allows for secure communications, which is very resistant to eavesdropping. Although, as I'm happy to get into in more detail, there is some question as to the practical utility of quantum key distribution over certain alternatives. In terms of quantum sensing. I think it is fair to say the US is in the lead, particularly when it comes to actual fieldable deployment. There's been a lot of scientific research for laboratory scale sensors by a wide variety of countries. But for the most part, those tend to mostly be only in the lab and not really fieldable, small, robust, lightweight, rugged, all the various requirements you need for an actual sensor in the field.

Computing, I think is the more complicated story. I think the very short version is still that the US is the world leader in quantum computing. However, there are different approaches toward quantum computing being pursued, for example, different qubit technologies, superconducting qubits, trapped ion qubits, neutral cold atoms, et cetera. And broadly speaking, I can go into more detail for each of these, but I think the short version is in most of these technologies, the US is ahead in demonstrated progress in the literature. However, in one of the leading technologies, superconducting qubits, the Chinese and the US are perhaps at approximate parity. The first really groundbreaking demonstration of superconducting qubits was I would argue done by Google at the end of 2019. Happy to go into more detail about that demonstration. And recently about a year ago, the Chinese demonstrated their own computer, which is roughly comparable in terms of technical specifications to Google's computer. So again, with various caveats, I think that would be rough parity in that approach. And the US is ahead in most alternative approaches. That's already perhaps a much more detailed answer than you wanted.

Yuval: This is perfect. I want to touch on quantum computing technologies as it relates to government direction and coordination. In the US, as you mentioned, various companies are working on different modalities of quantum from optical to cold atoms and superconducting and many others. It sounds from what you're describing that most of the Chinese work is on superconducting qubits. Is that, in your opinion, coming from above, or have they evaluated all these options and decided that this is the only one that works? What do the Chinese know that US companies don't know in that regard?

Edward: It's a good question. I wouldn't say superconducting qubits are the only technology that the Chinese are researching. If you look across all the literature they publish on essentially all of those different approaches. I think superconducting qubits is the only qubit technology in which China has demonstrated globally cutting edge capabilities. It's hard to say whether that was a deliberate choice to focus all their research efforts into that particular venue or whether they're working on all of them and that was the one that for whatever reason happened to be successful. We couldn't really conclude on that question. But there's no question they are working on a variety of different technologies even within quantum computing. An interesting wrinkle there is on their photonic approach. And again, this might be a little deep in the weeds, but one area of quantum computing in which the Chinese are very strong is photonic quantum computing, which I think overlaps with their high demonstrated capability in quantum communications.

But they're taking a slightly different approach, which is something called boson sampling, which is a somewhat limited form of quantum computing, which is not as powerful as a so-called universal quantum computing. And in this narrow sub capability of boson sampling, the Chinese are extremely strong and have demonstrated groundbreaking capabilities. They claim something called quantum supremacy using boson sampling last year as well, where quantum supremacy roughly means a quantum computer that is capable of performing some mathematical calculation, regardless of whether or not it's practically useful, which is too difficult for the fastest classical supercomputer in the world to solve. So again, it's a complicated story. And I wouldn't say that the Chinese or the US are putting all of their eggs in one technology basket,

Yuval: As much as we love the hardware vendors, the hardware is useless without software. Where does China stand with regards to software technologies to create efficient circuits, to create large scale circuits, and so on? Is that something that the report looked into?

Edward: We didn't do a deep dive into software companies or capabilities. For one thing, it's even more difficult to come up with benchmarks than it is with hardware. And benchmarking quantum hardware is already quite difficult. There are certainly a lot of companies, again, in both countries, and in many other countries too, not just those two, although we focused on those two countries, working on algorithms and quantum operating systems and user interfaces, both within small startups, and again, this applies the US, China, many other countries as well, Israel, as you may know. And there's certainly a lot of effort both by sort of the tech giants like Alibaba and Tencent, as well as in the US, IBM, Google, on the software side as well as the hardware side. So it's certainly an area of focus, but this report didn't do a deep dive enough to sort of make a comparative statement about different approaches by the two countries.

Yuval: When the subject of China comes up, inevitably people bring up industrial espionage. Do you feel that, one, were there demonstrated cases of industrial espionage in quantum? And do you feel this is any different than other areas, whether military technology or other type of industrial technology?

Edward: I am not aware of any specific instances of industrial espionage in quantum by any country on any other country. The topic of espionage, as well as collaboration internationally is a complex one. And there's a really good report that I'd like to recommend to all of your listeners, which is publicly available at quantum.gov, which was put out by the White House Office of Science Technology Policy, which is called something like Perspectives on International Talent in Quantum Information Science, which I think does a very nice job of balancing the two sides of the coin, because it's easy to focus on espionage and the risks of technology leakage. But we need to also keep in mind quantum technology is a very international area that is critical for the advancement of the science.

One of the statistics we gathered was the fraction of academic papers that were co-authored across countries. And we found that it was very high across the board in quantum technology, higher than in most technology areas. So this is an area of unusually high scientific interaction and collaboration. We think that's very strong. We want to maintain friendly, scientific ties with certainly our allied nations and potentially strategic competitor nations as well. So I don't want to say focus entirely on the risks of technology leakage across borders. That having been said, there have been in general evidence of high tech leakage of intellectual property, some legal, some illegal, to other countries, not all of them friendly to the United States. And that is also something that needs to be taken into account. So I'm not going to be prescriptive here and make any recommendations for policies, but I would absolutely recommend your viewers look at that report at quantum.gov and think about how we can balance both sides of that coin, the largely positive, but with certain risks, aspect of international collaboration.

Yuval: Let's tie that to the issue of export controls, which I think is one of the recommendations of your report. So if I'm a policy maker and let's say I'm convinced that quantum is strategic to the wellbeing of my country, then there are a couple of sort of knee jerk reactions. One is to say we need more funding for research because we need to create demand, we need to make sure that the companies are healthy, we need to make sure that there's research and manpower investment. That's all great. But then the other side, more on the defensive side, should we put in place export controls to make sure that technology that we deem is critical, just like military technology only stays or goes to places that we explicitly allow it to go to? What do you think about that and why are you recommending not to put in place export controls?

Edward: Yeah. So one thing you mentioned, which I would like to pick up on is the importance of maintaining strong funding for R&D. And that was another one of our recommendations in the report is to continue strong government support for R&D. We think while there's certainly the center of mass has shifted into the private sector in the past few years, still much of quantum is done at the open academic level. And it's not yet technically ready for the private sector to completely take the baton. And there's still a need for sort of basic fundamental research in the area, is one point. Regarding export controls, so our exact language in the report, if I recall correctly, were not to impose export controls on completed quantum computing and communication systems at this time. So all those words are kind of important, including at this time. So there are already certain export controls on certain types of quantum sensors. That's fully public and you can look that up at the Department of Commerce website.

So the reasoning behind our recommendation was frankly, it's not clear there are any applications of quantum computers yet. And even quantum communications, there are certain systems, but it's not really clear how impactful they are at this time. So we did not think that there was yet the ability to precisely tailor export controls to only apply to defense relevant applications because of the state of the technology. So we just don't think it's technically possible to have narrowly tailored export controls yet, because there are so many uncertainties as to what the applications will be and what the timelines will be for those applications. However, we did also say the federal government should be ready to change policy and impose export controls once the technology reaches readiness where those defense applications do seem to be in sight.

Another aspect to consider with export controls is many of these companies are still new and still have fairly low revenue, if any revenue, and their financial position is not necessarily entirely secure or sure moving forward. As we've just seen in the past few weeks, the stock market is rather unpredictable and a lot of tech companies took a hit recently in the stock market. So there's a concern that premature export controls might hit a lot of these, especially new startup companies, which might be relying on international sales as one of their relatively few sources of revenue. Now, obviously none of that is to put aside risks of technology leakage, but we just didn't think the technology was yet there where we could put narrowly tailored export controls. But again, we emphasized that as a provisional determination, which could change in the future.

Yuval: As we get closer to the end of our conversation, I'm assuming when you started writing the report or researching for the report, you had an open mind. You said, "I don't know what the conclusions will be. I don't have an agenda. I just want to report the truth as much as I can find it."

Edward: Absolutely. That's what we do at RAND. We do not write the conclusion before starting the research. Absolutely not.

Yuval: So what surprised you now that you've written the report? What do you find surprising that you weren't aware of or you didn't think was the case before you started?

Edward: That's a great question. I can think of several things. One thing came to the levels of government funding. So there's a lot of claims in English language media that the Chinese government is vastly outspending the US government in quantum technology. There's this figure of $10 billion, which is kind of thrown around a lot. And when we did a deep dive, we found we couldn't support that claim. We did not find evidence of $10 billion of investment by the Chinese government. When you go to the original sources, we found a wide variety of claims, everything from $42 million a year to $3 billion a year. Wildly conflicting claims, but we couldn't actually reach a conclusion as to whether the US or the Chinese governments are spending more on quantum technology, which is already kind of against, I think, the conventional wisdom. That was one surprise.

Another surprise was the drastically different structure of who is leading R&D in the two countries in the US I think it's fair to say the private sector is now at the cutting edge. A lot of the highest quality technologies are being produced by private companies. Whereas in China, there certainly is private industry in China and important companies. We identified several. But a lot of the major developments seem to come out of national laboratories, and in particular, one centralized national laboratory in the city of Hefei. So very different model there, much more sort of top down government driven in China relative to the US.

And the final one is there's always been a question of quantity versus quality publications. So people rightfully say you can't conclude very much by just counting publications or counting patents. You need to look at the quality. And that's hard to do at large scale. So I knew that China was putting out a lot of publications on quantum communications, or at least that was my anecdotal impression. But I was kind of curious how that would hold up when you adjust for quality. And we found that when you adjust for quality, you only look at very highly cited academic publications. The Chinese lead in raw publication counts and computing drops by a lot when you only look at the high quality publications. And China puts out significantly fewer high quality publications in computing than the US does.

But that wasn't the case in communications. We found that even if you only look at, for example, publications in the top decile of citations received, even after that correction, the Chinese have a comfortable margin, which if anything appears to be growing in recent years, over every other country in the field of quantum communications. So that does really seem to be a sustained technical lead, which is not just an artifact of putting out a lot of low quality publications. I think those are a few of my surprises, which weren't obvious going in.

Yuval: Edward, how can people get in touch with you to learn more about your work?

Edward: You can reach me over email. My email address is eparker@rand.org, and I'd be very happy to take any questions or comments from any of your listeners.

Yuval: That's perfect. Well, thank you so much for joining me today.

Edward: Thank you very much.

‍