Ladies and gentlemen, welcome to our session today on powering the next industrial era. We lack not for ambition here at Davos. You know where we are here to look into the future and see how we build that future. Am I right, gentlemen? I'm delighted to be here. I'm VJ Vitiswarren, the global energy and climate innovation editor of The Economist, and of great interest to me is the topic. As an engineer by training myself from MIT, I've been following this technology for a number of years as well as developments in AI, which I write about closely for us, and the linkage with energy. This really brings these topics together in a very interesting way. To my immediate left, we have Eric Schmidt, known to all of you, a former CEO and chairman of Google, and Bob Mumgard to my far left, co-founder and CEO of Commonwealth Fusion Systems. Please give them a round of applause to get them going.
They are going to give us a look into the future, but not really so distant future. Am I right, gentlemen? They've written a very interesting and provocative op-ed that's up on the World Economic Forum website. I encourage you all to take a look at it. Just to get a pulse of the room, how many of you in this room would say you're quite familiar with what fusion technology is? Hands up. Okay, this is great. At least a third, which is great to know. That means there's a two-thirds that are yet to be wowed by what's coming but also maybe skeptical or may have heard it's one of these classic 'always the energy of tomorrow' and we'll stay that way. So we'd like to update that view but also to point out what needs to happen to make it a reality. So Bob, maybe we can start with you. Commonwealth is the leading fusion company. I think you've just finished a round of funding that gets you to close to $3 billion in funding if I'm not mistaken. And you're pioneering some technologies in Massachusetts, but ultimately I think to build your first commercial power plant in Virginia. Can you give us a quick sense, quick update on where's fusion got to and why should we take this utopian technology of the future seriously now?
Yeah. Well, I think first for the two-thirds that don't know what fusion is, I'm going to give you the 30-second answer of what fusion is. And you can take it from me, I'm an MIT fusion guy. Fusion is a process that takes light elements, combines them together, and makes energy. Now, why is that interesting? Here on Earth, it's because if you were to build machines that create this process inside them, you would have machines that once they were built would make energy, as much energy as you want, for as long as you want, wherever you want, whenever you want, without consuming anything. So they wouldn't need to be part of natural resources of digging anything up or transporting it or waiting for the wind to blow or sun to shine. It would just be machines that made energy. And that's a big shift in how to think about energy because up till now we're sort of still in the hunter-gatherer mode of energy. And when you go from that to, hey, you make better and better machines faster and faster, cheaper and cheaper that have this reaction inside them that will never run out. We will literally never run out of fuel. And the fuel is available for everyone and there's no externalities, long-lived waste or runaway reactions. That is deeply sovereign energy. That is energy that could be used to do energy intensity at levels that are 10 times more than what we do today. So what would the world be like if you had 10 times more energy? And that is a big idea.
Climate change. There's also, it doesn't make any emissions which is useful. So there's no one to fight over the ins and outs. It's like it has no ins. So there's no one to fight over pipelines and supply chains. And there's no outs. There's no one to fight over what comes out of it.
So Bob, as Eric rightly pointed out, the local emissions, let's remember, millions of people a year die from the local emissions of the way that the world uses energy today, fossil energy primarily, but also on greenhouse gas emissions. So you have a benefit on that. But on the supply chain energy security you've talked about advantages and in terms of the opposite of intermittency, having reliability, firm power, that is a huge demand particularly from AI and data centers but industry as well. You see multiple benefits and this can be also distributed in fashion over time as you say and has the potential, as yet unproven, of being scaled, applying economies of scale so that you get cheaper over time. So this is the vision you're laying out. That's the vision.
Right. And so just so people understand why the enthusiasm, that's always been something of a vision although we've moved away from maybe the biggest of big science ways of doing this to now you're a commercial enterprise. You're backed by private companies including if Google was involved from the beginning but also doubled down on the next round. I think also is one of the offtakers for what you hope will be a commercial facility in the 2030s. Am I right?
That goes to the timeline. So you know the vision's been there since we basically understood how the stars worked. But what we've seen over the decades of research is we've been able to actually make this reaction happen here on Earth. We've developed classes of machines. There's different sort of architectures. And we've now recently just in 2022 in the United States built a machine by the national labs that made the fusion action happen at a level that made more energy than it took to create the conditions to make this very hot thing.
Is the ignition facility.
The national facility at Livermore. Exactly. And that's really a triumph of going systematically into the science of how this reaction happens and what the right conditions are. And that scientific journey has been very fruitful. But now what we see is we're starting to see that move away from just the science and become commercial. And this has happened in the past of all the other big industries we have today. At one point they were a science, like making aluminum was a science and now it's an industry. And fusion's going through that transition right now. And so you have the science has been proven. Yeah. No question. Can this be done? Now the questions are what's the most effective way to do it? How to scale it? How to get it out into the commercial market? You've got companies emerging. So we're one of them.
Let me stop you there. So you've laid out the case, I think, and hit some of the key points, but there may be some in the audience suspecting you're talking your own book. You are the leading fusion company. You've raised tons of money. But you got to the point about private sector. There are dozens of companies in fact, startups of various size, Europe as well as in the US and there's government efforts in China as well. We'll talk about that I'm sure in due course but the point is there's a global race. This is not one company, one technology, but rather the science is advancing. We're to the stage of engineering in fact advancing as well in terms of the financial models. And that's where I want to come to you, Eric. You've scaled multiple world-changing technologies in your career. Give us a bit of a handicapping of this race. How do you see, how excited are you? What's interested you in this topic now? What's the potential? And in practical terms, how does this move forward?
So first place, if you look at models of climate change, they're completely determined by what assumptions you make about clean power that's base load. So if you assume that fusion and fission are your only clean powers that are base load, which we can debate.
Some geothermal guys would complain but.
We can add geothermal in if you wish.
And again we'll see if that scales at the level of fusion and fission. We're going to have to solve the energy problem for the globe. Because as much as I'd like it, the fact of the matter is people are not going to use less energy. People are going to use more energy for all sorts of things including AI. So we have to solve this problem. You're well aware of the state of fission in the western world. Here in Europe most of the plants have been shut down except for in France. In America there have been a few starts. Most of them have gone bankrupt and so forth. This week by the way the national, the group that regulates this appears to have changed the standard by which you measure the emissions and the certifications of nuclear power. This is fission power and that's a welcome thing because it was really throttling the industry. But fundamentally we all need more power. Bob is a hero of mine because he took his PhD in basically in fusion stuff and in particular in tokamaks and he figured out that he needed a new set of essentially magnets. And so the first thing he did to get to where he is now, he solved a different problem which is a magnet. I'm involved with another company which is an alternative technology called essentially a form of pinching where you pinch it with a pulse and that requires a capacitor and again that company invented a new kind of capacitor.
So the interesting thing is that we're finally now not doing science, we're doing hard engineering, right? So your company's going to be successful because you spent years and all of that money to figure out a much more effective way of handling the essentially the race that goes around in the tokamak. You'll say it more accurately than I. So where are we is we're in the beginning of the engineering phase. Now, how much time do we have? You can look at climate models and decide. The core problem we have is that if you make reasonable assumptions, I'm going to use Q greater than one, which is a statement of excess power for you in 2030, 2031, some number like that. I noticed you didn't make.
You didn't make a claim. I had to go to make a claim.
Excellent. You heard it here, folks. You heard it here. That's next year. So net facility gain of Q greater than one would be incredible. The problem of course is you have exactly one and there is this huge scaling problem and the scaling problem can be understood as follows. As much as I want the private market to do it, we're going to have to have governments help. There are no global energy sources that did not have significant government funding and help and credit and backstopping and so forth to build it because the capital requirements are so high. It's very different from software.
So the way to understand the problem is first his stuff has to work. These other challengers have to work. We have to have the competition. There will be multiple winners. That's clear. And that's good. That competition is great. Then you have to scale. Now you start doing the math which is why I got interested in this. How many electricians does it take to build a nuclear power plant of fission, fusion, whatever? More than we have. More than we have in America. The demand for electricity if you just look at data center loads over the demand over the next three to five years is about 80 gigawatts. As a reference point, a nuclear power plant today is about one and a half gigawatts. Your demonstration site is.
400 megawatts. And that number has also been the same for a long time. I think you told me that the 400 was interesting because it fits right into the old coal-fired power plants. You can literally just take the steam engine out and put a new steam engine in. So it's connections, use the infrastructure, the same.
So you don't have to rebuild. This is a clever, again, a good engineering, good business decision.
But if you start thinking about the amount of power that we need, you're not going to get there by these kinds of techniques. You're going to have to have sovereign wealth firms investing large scale. So a team that I work with has built a plan for what it's going to take. Again, you're the physicist. I'm not. What are the hardest materials in a fusion plant?
Yeah, that's one of the interesting things. Like, you know, to build a new energy industry of any real significance, you're talking about building like 10,000 plants. Like, that's a good round number. If you do 10,000 plants, you're the largest energy industry. That's the number of 737s in the world. So, that's numbers. That's the scale. But most of the stuff is like steel, but it's steel. It's all in the right spots, right? There's a lot of intellectual property, a lot of know-how that goes into making the steel in the right stuff. There's some rare earths, but they're like dopants. They're small amounts again in the right spots. And so there's not a bottleneck on something you dig out of the ground. But there's a bottleneck on how do you get the stuff into the right shape. So manufacturing. So if you look at our supply chain, 1,300 firms, 30 countries and they are everything from people building that would be normally building reactors for oil and gas for say crackers. So big weldments and steel structures to people making chips for fast charging EVs. And so it dovetails with this existing supply chain, but you have to convert it. You have to build something like the size of the aerospace supply chain.
Does that call for something like a keiretsu model like the Japanese auto companies or even the tier one suppliers in the US? Is that what you're developing?
Yeah, it's basically what we're developing. I mean, today it's a lot more nimble than that. You know, you got to think of it, you're developing something that is like the early automobile supply chain where you're like going from ones and twos to like, okay, now we're going to make a bunch of them, but you have to have the same people, you have to have the same expertise, and you increasingly automate it as you go. There's pieces of our supply chain that look like that and there's pieces that look like someone who's got an existing supply chain, but they're sort of diverting one every once in a while and doing some retrofit to it.
And eventually it will look like a supply chain for building rockets, planes, and hopefully like an updated supply chain because you get to start over in some cases. So when you think about the role of generative design, the role of additive, like that's all very beneficial to fusion.
Eric, I want to pick up on.
So, if I can, I think based on that, I think the rhetorical question goes something like this. Do you fundamentally think that the state of fusion will end up with the equivalent of TSMC in Taiwan? Now, let's think about what actually happened. All of this technology was invented in America a long time ago. A bunch of clever people moved it to Japan and Singapore and then a particular founder who went to Berkeley became dissatisfied, went to Taiwan, convinced the government, they put an enormous amount of money and I think people do not appreciate that the Taiwanese success in semiconductors was direct industrial policy at a scale that in America at least people would blanch. It may take that. If it takes that then there will be nationwide winners who do it. It may not. It may be a little bit easier. Aerospace is not quite as concentrated. But also look at aerospace concentrated in geographies. Southern California for a long time. Seattle was a long time. So forth. You go to look at jet engines. Why are there only two providers of jet engines? And why has it been so difficult for China to build the third jet engine? Because the stuff is hard. And if you think jet engines are hard, trust me, fusion is harder. I know because I manage a rocket company.
So it is literally rocket science or even harder as you're saying. Just to put you up a challenging question to you Eric. Because you've thought a lot about the commercialization model. You've teased us that you've got a plan, a master plan involving I think sovereign wealth fund or some other manner of funding. Maybe we can talk a little bit about that. Certainly the people at Davos in this room and outside could well be interested in hearing that and being a part of it. But the challenge is this. You know the US got very animated about energy security, energy technology resilience a couple of decades ago in the wake of the oil shocks and we invested in the Synthetic Fuels Corporation. Europe in the wake of the Russian gas cut off got behind Northvolt for example as something of a local champion, European champion on batteries. We know the story there. It didn't work out. How do you guard against money poorly spent when it is directed in a particular technology? What's your sort of lessons from history to prevent that happening?
Well, it's what happens with these things is everybody talks about the failures of which there are plenty.
But you don't talk about the incredible successes.
Well, you mentioned airplanes already, of course.
But let's think about the GPS, right? The GPS essentially changed everything. Think, I mean again, Air Force satellites in 1970s. I would say that in America fracking is another example of such a success. There are plenty of successes as well as failures. The question is how do you get an industrial policy that does not constrict it and in general my view has always been that the right thing to do is to have low-cost credit, low-cost loans and so forth and let the commercial sector do it. Another way to say this is if you had the right essentially access to capital, you could build businesses in energy that are massive. Yeah. Now let's think about how important is it to get to fusion? It's immense. If you have enough capital, should you be in that business? Absolutely. Now, in practice, it's more capital than any one of us can raise. So that means we have to piece part it out. And I'm quite convinced in my understanding of how this is going to evolve that there will be a series of supply chains. There's the ones that you describe. At the end of the day, these reactors will have specialized parts and those specialized parts will be built probably by one supplier and the country that figures out where all those suppliers are and gets them in their own country is probably going to have a leg up. If I were to guess, I would guess Taiwan and Korea have the sort of attitude and structure to do this. Perhaps Japan, but if you think about the way Korea works and Taiwan works, they are able to do these things at scale. Think about the Korean shipbuilding industry for example. They're amazing at it.
Yeah. World class. Since we're talking about countries, I know you've done some deep analysis on how China's fusion efforts are going. So, it's really a question for both of you. There's no secret made by the Chinese. They're investing heavily in fusion. I think their leading effort looks curiously similar to what you're up to, Bob. Without casting any aspirations on how they got there.
Yes, exactly. But nevertheless, China has success in lots of areas, but they're also not good in some areas. The example of aerospace. I used to be our Shanghai bureau chief, so I followed the development of their engines very carefully, and they have not caught up with.
By the way, why was it so hard? Come on. It's just jet engines. Give me a break. 12,000 RPM. What's wrong with you? Why couldn't they do it?
Well, one top executive at one of the leading jet engine companies in the world explained to me, there are things that happen that are at the edge of supercriticality that don't follow the rules in the textbooks. We don't share that information. We don't publish. In other words, and I see you're nodding your head. You know, in plasma, you're finding this out day by day. That there are things that happen that don't follow past. And that makes it harder to cut and paste and cheat and copy. You have to invent and learn and learn by doing would be part of my answer.
