Everything You Need To Know About The Nuclear Energy Boom - EP 70 James Krellenstein

Back in the late 1960s and early 1970s, the US once had ambitions to build 1,000 nuclear reactors. These were all meant to be done by roughly the year 2000. We were going to be energy independent. It was going to be glorious. Richard Milhous Nixon was pushing for this agenda. I will let you guys know we did not build 1,000 nuclear reactors. The US Topped out at one point with a few more than 100. We're now down around 94 or 95 reactors. Over the last 30 years, I think we have built or brought online a grand total of 3 new reactors. People soured on this type of technology. Even environmentalists somehow mindfuck themselves into being anti-nuclear and It was— it's been a costly, depressing industry for a long time. But there is— there is new energy. I'm sorry, in, in around nuclear technology, we have a ton of startups building these things called small modular reactors. It's meant to, to bring down the upfront cost of building a nuclear system. You get these small reactors, you can place them right by a data center, by a town, whatever. Early days in many ways for this technology. Still trying to prove itself out. We also have a ton of fusion companies that are telling us fusion is finally real after all this time. So there's— in the Trump administration is, is trying to make all this stuff happen. And so it's exciting, but we need a reality check. And this is why we've brought in today's guest, James Krellenstein. He is the co-founder and CEO of a company called Alva Energy based in Massachusetts. James also happens to be remarkable. His knowledge of this industry is, is unmatched. He is going to walk us through the last few decades of, of the nuclear energy industry. What happened? What went right? What went wrong? Are these new technologies real? He's like, his brain is going to blow you away. Alva Energy is trying to make a kind of add-on to existing nuclear power plants. Where they can, they can add a turbine to these plants and pull more power out of them. It's, it's this relatively easy— I use that in quotes— way to, to give us, to make the most out of the power plants that we have. James is kind of a contrarian in some sense around all these startups and thinks a lot of them are making mistakes and are not going to pan out. But he's just— this is incredible. If you ever wanted to, if you just want to get up to speed on what's happened with nuclear technology over the last, call it like 80 years, why we are where we are, what's good, what's bad. You are going to find all of that in this episode. Hi, I'm Ashlee Vance. This is Core Memory. Let's get nuclear. Here we go. All right, James, you have a lot to live up to after my, my introduction. Oh my God. No pressure. I was like, I like to put people on the spot. I think people will be quite taken with your brain. I was thinking we've only met once and I was blown away by your encyclopedic knowledge of the nuclear industry, both the current state of things, the history. I then I've caught a couple other talks you've given. I mean, it's incredible the way you, you know this industry backwards and forwards.

Thanks.

Um, and so it's a huge privilege to have you here. I, I'm gonna, I'm gonna selfishly use part of this to inform myself and our listeners on, on kind of some very big picture stuff. And then we'll, we'll dig in deeper on, on these areas that, you know, so well. But yeah, I wanted to use this podcast as an excuse to give people a, you know, a, real knowledge about the state of the nuclear industry, the pros and cons. There's so much going on, being talked about with all these startups these days doing very different things. A bit of a reality check on what's going on. Obviously, the US and China is kind of top of mind. I see what China is doing with this technology. And, and so I want to— we want to come away with this, from this with a handle on a better handle on our world. Great. And then, and I also want to get into your your background a little bit.

Yeah, my background and the startup that I run.

Yeah, yeah.

Well, absolutely.

The startup you run. And then just after hearing you talk about a couple of the things. Yeah, I just want to get into your life and how you got here a bit. But okay, so let's, let's do this. Let's, let's set the very big picture for people assuming, you know, they're just casually familiar with nuclear. Again, I think they probably think not a lot's been going on in the US. Maybe they hear reports that China is building tons of new reactors. You know, and that might be the sum total of really kind of what they know, what they know. So, you know, I mean, I ran some numbers before we did this. The United States, according to the internet, is still the largest, has the most nuclear reactors. 93.

Yeah, I think it might be a little higher now. I think it might be 94. But yeah, the United States is the largest producer of nuclear power by the number of gigawatt hours or terawatt hours generated. I think what's important to recognize is that, you know, although we are still the largest, we are on track to fall behind China pretty soon. And part of the reason for that is, with the exception of Vogtle Unit 3 and Unit 4, which came online in 2023 and 2024 respectively, the last nuclear power plant that we started construction in the United States that actually turned on was Sharon Harris Unit Number 1 in North Carolina that started construction in 1978.

Yeah, I was shocked. So when I was researching for the show, I mean, you mentioned that Vogtle is two reactors in Georgia. Yeah.

And but two new reactors, two new reactors, two existing reactors there, Vogtle 1 and 2.

The internet was claiming so they had three. We'd only built three in the last 30 years.

So probably three have come online. Okay. And that's probably Watts Bar Unit Number 2, which comes online in 2016. But that plant started construction in 1972. So it started construction in the '70s and then literally finished in 2016. But I think, you know, for what we're talking about here, you know, Vogel 1 and Vogel 3 and Vogel 4 started construction, at least in nuclear construction, in 2012 and 2013. And they completed in 2022, 2023, and 2024. But prior to that, we hadn't started a new nuclear build that actually finished until— since 1978.

Yeah, this is insane.

And older than I am. In fact, my father, who's a nuclear engineer, he— when he graduated from nuclear engineering school, he worked on Sharon Harris, and Sharon Harris was— had already started construction when he graduated college. Just to give you an idea, that was the last one we did before Vogtle 3 and 4.

And so, so 3 decades. Very, very little happens. At one point, the US was talking about we would have 1,000 nuclear reactors by the year 2000.

Yeah. So clearly President Nixon.

Okay, so we went off track there. And then just to do the numbers real quick, so US roughly 94 reactors, France around a little bit more than 50, almost the same as China, somewhere around like 56, 55.

And China has right now about 40 reactors, give or take, under construction right now.

Like, yeah. So they're about—

they're, they're going to, they're going to get ahead of us probably probably in the next 5 or 6 years.

Okay. And then, and then we'll just tick off the other ones. It's Russia, South Korea, India, Canada, Ukraine, UK, and Belgium with 7. Good for Belgium. So, you know, I mean, the numbers, they sort of shocked me in how I knew nuclear was stuck. I did not know it was like this stuck in the US.

So there's a couple of things to, you know, I think a lot of people say nothing happened. I wouldn't say that. One of the— nothing's happened since the '70s. One of the things that is weird is that although we've kept the basically the number of reactors the same, or actually in some cases actually going down, we've shut down a good amount of plant. At one point there's about 112 nuclear power plants online in the United States. We have actually increased the amount of power that those plants are producing a lot. And that's primarily because of something that we'll get into, which is called capacity factor. So and this is something that any power plant has. What a capacity factor is, if you had a a 100-megawatt power plant and it was making 100 megawatts for 365 days, right? We call that 100 megawatt days that it generated. That would be 100% capacity factor. But power plants, like anything, don't work all the time. They have to go down for maintenance. Maybe they don't need it all the time. So a lot of power, you know, if that power plant was averaging 50 megawatts, even though it had 100 megawatt capacity, the capacity factor would be 50%. What happened in the 1970s, although we didn't build that many more plants or really any more plants after '78 except for Vogtle 3 and Vogtle 4, the capacity factor, the ability for us to operate those power plants a lot and generate power increased dramatically. In 1975, the capacity factor of a nuclear power plant was about 55%. Today it's over 90%. So even though we didn't build that many more power plants, we actually got roughly double the amount of electricity out of the same power plants by learning and building technology to take those exact same power plants and run them much more reliably. And that's really key to where nuclear is today.

Is that the case just for the US or that's worldwide?

So it is worldwide, right? I mean, you know, nuclear power plants, we forget that we had never— man had never really captured the strong nuclear force, which is what nuclear power ultimately is, is powered by and tamed it until December 2nd, 1942, when Enrico Fermi and Walter Zinn built a nuclear reactor literally in an abandoned squash court in the University of Chicago. And before that, sustained, self-sustained nuclear fission reactions hadn't happened on Earth for about 2 billion years. And you know, from that point, it took us a long time, but actually not that long of a time, but it took us a couple decades to figure out how to take this new source of immense energy and figure out the engineering systems that are necessary to actually convert that into useful work and then into electricity, electricity, and then how to run that reliably. So we learned a lot of how to run power plants over this time, even though we weren't building them, we actually built a huge amount of technology to be able to actually run them cheaply and safely and at very high reliability. And that meant that we have kept the amount of power that nuclear is providing in the US at around 18 to 20% pretty constantly for years, even though we haven't built that many more power plants.

Okay. We've talked about Vogel a couple of times. Sure. And I just want listeners to know we are going to explain that in a second. But before we do that, so just going back to China for a second. So you laid out what's been happening in the US.— we talked about China's going to double roughly their number of reactors. The— what, what is their strategy with their reactors? Is it, is it just kind of mimicking already what they're doing and what the US has done, just building more of it? Yeah.

So what's really interesting about the Chinese is that Chinese right now have a very, I would say, deliberate and engineering-first approach to nuclear power. Their bread-and-butter nuclear reactors that they're building, I believe 38 of right now, are actually direct descendants of American nuclear technology.

It's this AP—

well, they are building a version of the AP1000 called the CAP1000 and the CAP1400. Chinese AP1000, Chinese AP1400. And that is a Westinghouse design. That was the reactors that we built at Vogtle 3 and 4. And the other reactor type that they have built is called the Hualong One. And the Hualong One is actually a derivative of a French reactor type called the M310 that is derived from an American nuclear power plant in Beaver Valley called, called the CPY. Right. So we took the Westinghouse 3-loop, we kind of gave it to the French. The French standardized it, built about 20 of those reactors, maybe a little bit more. And then they exported that reactor technology to China and the Chinese indigenized it and standardized it as the Hualong-1. So these are both pressurized water reactors. They're both uranium dioxide fueled. They are like the power power plants that we have in the United States and France. They are not the advanced reactors that a lot of the startups are building.

Okay. And I've seen you talk about this, so I think I know the answer. If memory serves, the story of how they got the Westinghouse AP1000 is perhaps not America's finest hour.

Maybe not Westinghouse's finest hour.

Yeah.

But, you know, in the, in the mid-2000s, Westinghouse in engage in pretty significant technology transfer to the Chinese to build, uh, to transfer the AP1000 technology over to China. And actually, the first AP1000s, although we started building 4 AP1000s in the United States, we only finished 2. The first AP1000s that ever came online in the world, these Westinghouse American designs, came online in China. Sanmen Unit 1 and Unit 2, and Haiyang Unit 1 and Unit 2.

Do.

So the Chinese— we did take kind of in some ways the crown jewels of some of America's nuclear program. We did transfer to China, and we allowed the Chinese to really actually, you know, do tech transfer and standardize that design and make it their own, which they did.

So their workhorse, for all intents and purposes, is based off this Westinghouse design.

Yeah, I would argue that their work— their main workhorse is not the CAP-1000. They're building you know, a couple of them, but they're mainly building this Hualong One. The Hualong One grandfather, if you will, or grandmother, I don't know what the gender of it is, but the, the parent, the grandparent of it, to be woke, is the, is the Westinghouse three-loop design that was built in Pennsylvania at Beaver Valley and at North Anna in Virginia. And that was transferred to to France, which then transferred it to China. And we sort of got this sort of nice lineage of technology. And maybe it probably— unless people know what I'm talking about, I mean, should it make sense for I— what is a pressurized water reactor? I feel like I'm kind of tossing out terms like a nuclear person and not actually explaining what the hell it is.

No, you know what? Let's hold that for just one second and then we'll get into some of the different reactor types.

All right.

What do we do at Core Memory? We cover innovative, fast-moving, forward-thinking companies, which is why Core Memory is sponsored by Brex, because Brex is the intelligent finance platform for many of these companies. 30,000 companies from startups to the world's largest corporations rely on Brex's technology for their finances. They've got smart corporate cards, high-yield business banking, and expense automation tools that are fantastic. I hate doing my expenses and Brex's AIs and software run right through those expenses, figure out where we're spending money, and take care of so much stuff for you so you don't have to waste your time on it yourself. Go to brex.com/corememory to learn more and just, you know, get with the program. Let's get going. Let's get out of this archaic finance software and move toward the future. CoreMemory and Brex. While we're on China.

Sure.

And, and I like that we got a woke nuclear—

it's a rare thing.

This is a rare, rare podcast moment. The— do we know roughly because I— well, we talked about how the— not much has happened in the US, but do we have any sense of comparison of how much it costs China to make a new reactor versus what it would cost in the US and how quickly one would build in either country?

We built— me and a couple of colleagues my company at MIT and Argonne National Lab and Idaho National Lab. We wrote a whole research report on this exact topic. But, you know, the Chinese can make the reactors a lot cheaper right now. I mean, labor costs are a lot cheaper. They also have a much more established workforce to build large projects like a large light water reactor.

The—

there's the caveat here, though, that actually when they tried to build the first AP1000s, they actually took about the same time to build in China as they did in the United States. So I think a lot of the common sort of bullshit reasons why you hear why China moves faster is not necessarily actually the real reasons, right? Even, even if you just took the American reactor design the first time and tried to build it in China, it built just as slowly as it built in the US. It was cheaper, but it built about the same timeline. The Chinese right now can build on average a— on the median build times about 5 years from beginning of nuclear construction to commercial operation. When they tried to build the first AP1000, Sanmen Unit Number 1, it took about 9 years, 4 months. So more than 2 standard deviations longer when they were building that first American reactor using the American sort of building techniques rather than what it builds on average in China. And we are seeing even that AP1000 design, though, dramatically accelerate in timeline as the Chinese have standardized the supply chain, standardized the design, and made it constructible and optimized to actually be built into the field.

And I mean, I guess there's obvious answers here, but you're the expert. So China's committing to nuclear so dramatically because, you know, if you look writ large, it seems like they're going whole hog on nuclear and solar, moving away from oil, moving towards cleaner technology. Is it— is it more complicated than that? Or—

Well, I mean, just pulling back for one second about the Chinese energy policy, you know, with the exception of coal, right, they do not have almost any fossil resources. Right. They don't have particularly strong oil or gas reserves. So, and you know, we don't just look at the headlines of the newspaper today, right? Strait of Hormuz is basically shut down. That is really going to be strongly impacting China, right? In the global LNG, liquefied natural gas market and the petroleum market. So China is viewing nuclear energy alongside solar and batteries and some wind and coal, by the way, as part of their, you know, energy independence project that is really necessitated by geography. They just do not have the fossil fuel resources necessary to, uh, power their economy except for imports, which they do not want to rely on. So the Chinese are viewing these resources as a need-to-have source of energy for their economy. So they are going whole hog, I would argue even much larger and larger into solar and batteries than they are into nuclear, but they are building a lot of nuclear power as well. And I think actually in the recently announced 5-year plan, you're seeing actually a re-emphasized emphasis on nuclear and a little bit less on solar, although there's still a lot more solar than nuclear, relatively speaking. So that is something that is very similar to what the French did in the 1970s and 1980s after the Arab oil embargo in 1973. The French, led by Prime Minister Mesmer, basically committed to the idea that we're going to nuclearize the entire electricity sector in France. No more coal, no more oil. And what they were able to do is in a matter of a decade or two, you know, they basically decarbonized effectively the entire French electricity grid, increased the amount of electricity that France made by 80% simultaneously, and got off completely off of oil for power generation in, in a one-fell swoop, which really gives you an idea of the power of a well-executed nuclear scale-up plan. The reason why President Nixon in 1973 had that speech, I believe, in November of '73, where he committed to 1,000 reactors being online by year 2000 was because of the need for energy independence. Nuclear power is incredibly energy dense. It allows just a truckload of fuel to power a city the size of San Francisco for 2 years. And that is a transformative technology for geopolitics and energy independence, which—

okay, which leads us to all kinds of questions. I mean, so, so France, I mean, sounds great. And but they, they really stand out as especially for the size of the country, as, as unique in pursuing that approach. If it makes so much sense.

Well, France, I think France, it's important that to realize, I mean, I was just in Paris yesterday, right? The and maybe even meeting with some of the French nuclear industry there. The, the, the French have a really interesting story to tell. About how they actually did that. And there's problems, let me tell you. Boy, are there problems right now in the French nuclear industry. I don't, I don't mean to suggest, and certainly if you look at their recent bills, not particularly great. But the French did a really, really interesting strategy that maybe the United States and other technologists could learn from in that the French actually, even before they went on this MESMER plan in the '70s, they had their own indigenous advanced reactor technology called the UNGG, right, which was a high-temperature gas-cooled reactor reactor similar in some respects to a lot of what the startups are building today, right? They were gas-cooled, graphite-moderated, did not use light water. They used natural uranium rather than enriched uranium. And the French made a really bold decision in the late '60s and early '70s to abandon the so-called advanced reactor technologies and focus on the tried-and-true proven pressurized water reactor technology that was pioneered by Westinghouse. And their approach was really quite pacing. What they basically said is, we're gonna go over to the United States, we're gonna go to two nuclear power plants, one in Pennsylvania called Beaver Valley, the other one called North Anna in Virginia, and we're just gonna copy it. We're gonna get licenses to it. We're gonna create a joint venture with Westinghouse called Framatome, which literally stands for French American Atomic Constructors. And we're just gonna take that reactor design, standardize it and build it over and over and over again, dozens of times. And that was the base, you know, the foundation of how the French literally built their nuclear dominance in Europe, right? They said, we're going to, we're going to abandon the advanced reactors. We're going to, we're going to look at a really proven technology that can generate power reliably and economically and just build the same thing over and over again, minimizing any changes. And after they built around 23 of them, they went in and they said, okay, we're going to now actually do slight improvement, we're going to go down to Texas, kind of take a reactor called South Texas Project. We're going to standardize that out and build another dozen of those reactors. And then they did some— a little bit more improvements on the N4 reactor. But that was really their approach. It was do not get sort of distracted by really, really fancy technology. Look at what works and what can reliably generate power 24/7. Every day and, and really, and every week, and, and really standardize and get good at building that, build the supply chain and vertically integrate it.

So I mean, this all, yeah, seems like a smart plan. Seemed like you're saying there's issues now, but why, why did nobody else mimic what, what France did?

So I mean, I think there's two issues here that, um, that play, play into this. The first is that the French are, were rather unique in that they not so unique, but they had a national utility called Électricité de France, EDF. And so they had one utility for the entire nation. And so EDF literally said, we are going to consolidate all of the project delivery capabilities, all the architecture, engineering, all the construction capability in a single institution, and we're going to keep on building the same reactors at sites all over from Fessenheim to Bugey to Tricastin. We're going to just literally, literally have one team that is doing all of of it. And that was something that was not possible in the United States or most other nations, which might have dozens, or in the case of the US, thousands of different utilities. So really could be coordinated under a single approach. The second, I think, was a sort of very French, especially in the '70s, '80s, sort of Fifth Republic sort of time politically, of having these grand engineering projects. You know, it wasn't just the nuclear power plant, it was also the TGV, the high-speed rail. It was also the Concorde, right, with the Right. Really, this idea of— Mesmer said, you know, in France we don't have oil, but we have ideas. And there was a real idea that this would be a showcase of French engineering excellence and a commitment through multiple successive governments to commit to fulfilling that, that role. And I think the final thing— I know I said there's two, but there's probably three— is they really did not have very many fossil fuel resources in France. They still don't. The United States does have a lot of fossil fuels, right? Huge amounts of coal, as everyone knows. We have huge amounts of natural gas and huge amounts of oil. That's why we're the world's largest oil and gas producer today. And so they had to, in order to meet load without being dependent on foreign countries, they really had to actually figure out how to do nuclear and do it well.

And okay. And wow, where did that spirit go? I just was thinking of Arianespace telling me that reusable rockets are impossible.

Yeah, no, I mean, I mean, let me just say that that French spirit has maybe, and it's up to someone who knows the French je ne sais quoi better than I do. I don't necessarily say that's where France is today, but France certainly was that in the '70s, '80s, and '90s on these heavy engineering projects, even Ariane, and even the ESA and Kourou, and all like the, you know, like that was all sort of this weird French, you know, really French-led sort of engineering attitude. I mean, and you could argue Airbus sort of, you know, with Bernard Ziegler really taking Airbus from like a kind of EU make-work project into really the cutting edge of, of wide, of, of commercial airplanes. It was also there too. I'm not so sure it's there anymore. Yeah, it's up to someone else who studied that more than I have.

I've been to Karoo where the French launch from Guiana several times. It's quite an experience. It's actually very cool. So, okay. This is like a very loaded question. Well, it's not. I mean, it's just got— it's too big, maybe. But okay, so you're saying Nixon says we're going to build 1,000 reactors in that year.

Let me just be clear. In 1973, U.S. utilities order more than 100 new nuclear power plants.

Okay.

So people, right, 1/10 of it. Yeah, I think it's already happening. But Nixon— Nixon's making the speech in November that he's making the Project Independence speech. But in 1973, U.S. U.S. utilities order 100 nuclear reactors.

And then, okay, so everyone always points to Three Mile Island and then Chernobyl and then just fear about nuclear waste. And then you weigh that against where we ended up with, with this big push around climate change where nuclear seems to make a lot of sense. And yet, you know, so we've got, we've got things against, we've got major things for, but then not much happened. So what, what happened between when Nixon makes—

So Three Mile Island, I think, was the cherry on top of something that was already happening, right? So, you know, as I noted before, the last nuclear power plant that had been ordered was Sharon Harris in '70— or started construction was Sharon Harris in '78. Three Mile Island happens in spring of '79.

Okay.

So even before TMI happens, Three Mile Island Unit 2, the meltdown there, we saw a real slowdown completely of nuclear construction orders. And that really is starting to happen in the '70s. That there's a dramatic slowdown in nuclear construction because we had massive cost blowouts beginning to happen at nuclear power plant construction in the United States. So, um, there's a couple factors going on. One of the biggest things that happens is, you know, ironically, the, um, Arab oil embargo coupled with some other macroeconomic factors leads to really high inflation in the United States. And in order to reduce that inflation, Paul Volcker and the Federal Reserve really, uh, cut, you know, knock up interest rates a lot. So a nuclear power plant is a very capital-intensive endeavor. It takes a huge amount of capital to build. It has very low operating costs, but very, very high capital cost upfront. And you generally finance that by borrowing money. So even without any change in the underlying power plant, right, even what we call the overnight cost is that's the cost of building the plant as if you could build it overnight and therefore the cost of money is basically zero, right? Even that was constant. You know, in some cases we saw the total price of the power plant double or triple just because the cost of borrowing money got so high. So that's the first thing that happens. The second thing is, is there was huge labor shortages going on in the 1970s that were impacting both nuclear power plants and fossil power plants at the same time. You know, in between 1967 and '75, we saw fossil power plants more than double in power and price just for the same, you know, unit megawatt. Kilowatts being produced, and nuclear power plants actually went up by about fivefold. So more than that, but it was in an overall environment of real, of real increases in power plant costs just because the cost of welders, cost of labor went— shot up through the roof and the cost of commodities like metal and concrete went up as well. And then the third issue was, is we had real regulatory issues that were starting to arise as the nuclear power industry in the United States scaled. And as we had multiple different vendors and different designs being built at different sizes all throughout the country, built by different utilities and different architect engineering and construction companies, and that coupling with a pretty unstable regulatory environment really started driving the cost of these nuclear power plants way up and the time it took them to complete. Way longer. And so power plant— so utilities were canceling new nuclear power plants left and right.

And then, okay, so Nixon sets this ambitious goal. All these factors play into slowing it down. And then at some point, people must step back and be like, well, we haven't done any. This is quite a difference, right, between this huge goal and then the whole thing grinding to a complete halt. Yeah. Yeah. Were there— at what stages were there alarms going off or America was just just trucking ahead.

And I don't think there was, um, I mean, there were alarms, certainly. There was a lot of attempt to revitalize the nuclear industry. I think one of the big things that happens— let's pull back for one second. I mean, we're kind of going all over the place here. I mean, the big thing that actually is really driving the change in power plant construction at this time is just power usage. The growth of power usage slows way down in the United States between 1945 and 1975. '73, the United States, the post-war boom we can call it, uh, you know, power use in the U.S. was growing extraordinarily rapidly, far, far faster than it is today. We were on average doubling the amount of power that the United States used every 8 years. So the just demand to keep the lights on at a power company was so huge that you were just building power plants left and right. And we start really after '73, a couple things are happening. There's a little bit slow growth in industrial industrialization, we see a lot of efficiency happen into the power, into various technologies. And we start also seeing that a lot of people who are getting air conditioners, getting electric stoves, you know, electrifying the house— that happened. So what happens from '73 to about the '90s? Power grows from doubling every 8 years to doubling every 25 years. So just power growth was slowing a lot down. So just then, so a lot of these power plants, these nuclear power plants, were canceled in this time because the load, the is the increased demand for electricity that they thought was going to be there wasn't. Um, and that meant that just the utilities themselves changed inherently in what the utility was doing. You know, in the '50s, '60s, and '70s, if you were at a utility, you were building power plants. That was a huge part of your business because if you didn't do it, the lights wouldn't be on and your customers wouldn't have power because you just needed to keep up with that demand. But starting as we slowed way, way down in the, in the mid-'70s, the capability of these utilities to do that sort of work also started atrophying a lot and they just didn't need to do it as much. So there wasn't as much alarm as you would think. Obviously if you were in the nuclear industry there was, but just nationwide, 'cause we thought, hey, like, you know, we don't need the power in the same way that we did it in 1965 and 1967 and 1970. 1968. And we had a mature nuclear industry at that point. You know, in the '50s and '60s, government policy was really correctly focused on how do we launch, how do we incubate a nuclear industry? How do we get big companies— excuse me— like General Electric, Combustion Engineering, Babcock Wilcox, and Westinghouse, the Big Four, how do we get them to set up really solid nuclear power businesses? And they did do that. I mean, we ended up building over 100 nuclear power plants in this country that powered 20% of the nation for a very, very long time. That mission was successful. And then when the market forces started slowing down on power consumption, with some of the predictions about where oil production was going to be, like the Hubbard Peak turned to be not as true as was thought, there was a little bit of a nonchalance that occurred of saying, okay, we're not building the reactors, we're still finishing a bunch of them, but like, look, if there's not customers, we're not gonna order quarter of them. And so even you saw even government utilities cancel nuclear power plants like TVA canceling Yellow River and other of the TVA plants that were abandoned mid-construction. So and then the final thing was the end of the Cold War, right? And by '91, right, you know, our competition on nuclear technology with the Soviets sort of evaporated for its raison d'être. And we started seeing demobilization of some of the more experimental research technology like Experimental Breeder Reactor 2, EBR-2, out in Idaho, as well as abandoning, you know, Clinch River Breeder Project and that sort of next-generation nuclear technology.

Okay. That's—

there's a lot of reasons, but that's— there's a lot of history here. And I guess we're doing a historical deep dive.

We are for a minute, just because I want to get us to the present day. But I want to understand—

and we haven't really talked a lot about the tech, and maybe that would be actually kind of an important aspect of this as well.

All right, pod friends, I am here to tell you about SendCutSend. They are a manufacturing phenomenon. Whether you're working on your own car or you're a giant automotive company, a rocket maker, if you need metal cut, machined, or bent, you're going to SendCutSend. They'll do it for you right, and they will do it for you fast. Go to SendCutSend.com/corememory for a 15% discount on whatever you're trying to make. We love SendCutSend. Go visit them today. And then just, so just one more kind of historical question. And again, I mean, it's a massive open-ended question, but you know, you're talking about the end of the Cold War, you know, around this '91 period. It's not that long after I think that climate change starts to bubble up more in the public consciousness. I forget when Al Gore's movie came out, but it was somewhere around like '95, '96, I I don't think so. I don't know.

Maybe it was. I'm not a huge, uh, Inconvenient Truth fan.

Okay. I hate to tell you. Okay. No, no, no. I mean, it just, you know, it was like a, it was a moment, right?

Yeah, I guess so.

Let's see.

Yeah, no, I mean, um, I actually, I don't, you know, I think the, the, the first nuclear renaissance more than anything was spurred by the, um, probably by 9/11 and stuff.

Oh God, I was off. Yeah, this is, uh, 2006.

Yeah. 'Cause I remember I was seeing the that, and I was a little baby.

Okay. I mean, my question was, was kind of, as climate change starts to, to get a lot of attention, and yeah, you know, this is again a big broad question, not like I'm the first person to think of this. I was just always struck, I mean, for a while it was like the people that you would think would kind of be into clean energy, yeah, still like anti-nuclear. Why was that not enough? Well, okay, the hump.

I don't want to— you know, you want to talk about like the history of the anti-nuclear movement. I mean, that's a, that's a whole podcast episode.

Yeah.

So, but you're absolutely right. The environmental left, let's call them the environmentalists, were, were very anti-nuclear up until very recently. And even now it's not as, you know, if they actually were looking at the engineering and physics, of course they would be very pro-nuclear. But a lot of this, you know, comes out of the fact that in the '70s and '80s and '90s, there was a huge anti-nuclear movement in the United States. The States, um, that was part of the environmentalist movement and part of the anti-war movement, right? So a lot of the opposition to nuclear energy was dealt with the conflation of nuclear— civilian nuclear energy with nuclear weapons. And, you know, obviously nuclear power technology itself, especially light water reactor technology, is a pretty low proliferation technology. It's not really the way— no one's ever used a light water reactor to, to get to a bomb. You can't really. It's very, very difficult to to do so. Um, but a lot of the same industrial players were involved in both the weapons complex, uh, in the arms complex, as well as in the nuclear power complex. So that led to a lot of the anti-nuclear movement conflating the two, coupled with a sort of NIMBYist, like, uh, not in my backyard anti-nuclear vibe. Um, and they were very effective in the '70s. I mean, they killed probably a good amount of nuclear power plants, were canceled ultimately because of environmentalist opposition. Ironically increasing the carbon emissions and, you know, deaths due to air pollution really significantly in the United States. And then what, what occurred, I think, um, is that that same group of people became highly concerned about climate change, um, without necessarily being physicists or engineers and not understanding that actually we're not gonna, you know, be able to do this without solar or wind or batteries. So they, even though they were hypothetically ostensibly saying that they're very concerned about climate change, they ignore the single technology that has been most effective at decarbonizing entire societies, which is like nuclear, nuclear energy, like what it did in France. And I think there's also, if I may, may opine, one of the interesting things about nuclear power is that it kind of is a little cheat and it's a good cheat in my opinion on climate change. You know, a lot of the climate change advocacy, in my opinion, was sort of centered around a sort of moral crusade. Like, you know, you have to, you know, not turn off— turn up the AC and suffer in order to save the planet. But what's cool and great about nuclear power is you can use as much fucking energy as you want. It just doesn't emit carbon dioxide. It's kind of like when you go to Paris, no one thinks, oh, you know, the City of Lights, this is a decarbonized green place, right? You go to Germany, they say all the fucking shit about how decarbonized they are, and they're making 7 times the amount of carbon dioxide per unit of power that the French are. But in France, it's just much cleaner. And that's what's great about nuclear power, is, is that it lets you decarbonize without you thinking about decarbonizing. You don't change your life, just the power is cleaner. And I think that also was a problem for the environmentalist movement, who often, as we've seen, seen, like to conflate environmental objectives with political and social and moral objectives that actually are not necessarily related to reducing carbon dioxide or greenhouse gas emissions.

Yeah, yeah, uh, it's bonkers sometimes.

Yeah, it was really fucking stupid. Let's just call what it is. And like, then you see, you know, the, the people in Germany, right, who led by the allegedly the people who are most concerned about carbon dioxide emissions shut down all their nuclear power plants, killing thousands of people a year because of the air, the excess air pollution that was created because of the continued reliance on coal and made their grid incredibly dirty from a carbon dioxide and greenhouse gas perspective. So one does really have to ask, are they really concerned about climate change or are they using climate change as a fig leaf for other political objectives. And I think the empirical evidence overwhelmingly says the latter rather than the former.

Yeah. Okay. So this has really helped bring us— this is— I wanted to do this big picture, kind of get us to this moment in time, because if you're in the United States, nuclear is cool again in a lot of ways.

So, yeah, I mean, I think it's— it is cool again. And I say that as a nuclear company startup founder. I'm not going to— I'm not going to argue agree with that. But, you know, there was a first nuclear renaissance that happened early in the 2000s in the United States that led to Vogtle being constructed and finished, and Summer 2 and 3, the other nuclear power plants that we started building and didn't finish in, in the 2000s. And that was actually— I think, you know, there's a lot of factors going on, but it was a lot of times actually, um, I'd argue September 11th really had a big role in that, uh, leading to the Bush administration passing the Energy Policy Act of 2005. And that's sort of the foundation of the moment that we're in. So to try to get off oil again, we'll get off oil and get off gas. Yeah, I think we forget that before the fracking revolution, you know, before the Rice brothers and others really revolutionized the American— sorry, energy industry, what occurred, you know, natural gas prices were really quite high in the US. In the US and coal. There's a lot of coal, but people don't want to use coal. We were importing most of our oil at that time, or a good amount of our oil. And so there was a real concern, hey, if we're going to be able to power our economy and not be overly reliant on the Middle East, which was an unstable region now, and right now an unstable region as well, you know, we're going to have to figure, figure this out. And so the Energy Policy Act of 2005, signed by the Bush administration, created a huge amount of funding and new apparatuses to facilitate the building of nuclear power, as well as do a lot of the foundations for the fracking revolution as well. Right. Like the Clean Water Act exemptions that were part of what allowed fracking to really take off.

Well, okay. So I promised people we'd explain Vogel. And so maybe this is a good spot. I mean, tell us what happened the last time we tried to have a nuclear—

Well, so there's a couple of things that really happen there. The first is that they create an office then called the Loans Program Office, now known as Energy Dominance Financing. And what, what LPO— that's what—

I'm sorry, Energy Dominance Financing.

Energy Dominance Financing is an office of Energy Dominance Financing, formerly known as the Loans Program Office.

That's amazing.

You know, it's dead name, as they joke, is, is LPO, is the Loans Program Office. This is a Trump rebrand. Oh, it's powerful. I will, I will let others opine on, on the branding. Not my department. But basically what that LPO/EDF was created to do was to solve that financing problem. So in a nuclear power plant, as I talked about before, even if the cost of building the plant, you know, the cost of the concrete, the parts, the, the workers, even though that all remains the same, if you go from a cost of capital of 5% to 9%, everything else being the same, the cost of the plant doubles. So what the idea of LPO and EDF was, while the US government can, uh, it used to be able to borrow cheaper, but still could borrow money very, very cheaply. What if we basically allow, um, utilities to borrow at basically the rate with a little bit of an adder of where the US government borrows money? And that way we can just reduce the price of the power power plant and the price to the ratepayer just right off the bat by, by, uh, decreasing the cost that it takes to borrow money. The second thing is that they basically, um, provided funding for next-generation reactor technologies like the AP1000, like the ESBWR, which never got built, to get through the licensing and final design process. They've put hundreds of millions of dollars in a consortium of utility. They created a, a consortium called NewStart, like nuclear start, I mean, it's kind of nuclear industry is not very good at branding, but New Start, it's like a, it's a funny pun. And then basically New Start was going to cover some of the cost of getting that design certifications done at the NRC and then also getting the licensing applications through for the utilities. And we forget that out of that process, the United States licensed, fully licensed from a regulatory perspective, green flag, 17 gigawatts of new nuclear capacity as part of that process. Right. Are, although we only started construction on Vogtle in summer, there are fully licensed sites all over the country. There's places in Florida, there's 4 units in Florida that could be built that, you know, there's nothing left from a regulatory perspective to start construction. Same thing in South Carolina, same thing in sites in Michigan, in Texas. There are sites all over the country that are ready to go from a regulatory perspective, have everything, the environmental impact statement done, the combined operating license from the NRC, the evacuation plans, the, you know, public hearings are all done. That, that actually was successful in getting the regulatory stuff ready to go. But we ended up not building any but 4 of those 17 gigawatts and we only finished 2 of them.

And in Vogles, Georgia. Yes. Summers, Washington.

South Carolina.

South Carolina. Okay. So let's, let's break those down then for a second. So, okay, so we're kind of—

the government is, is pro again 20 years ago and, and some work gets underway and the government You know, one of the things that actually begins happening is nuclear power, you know, historically has been something that the Republican Party supported and the Democratic Party opposed. And that is still— that dynamic is still around, but it's weakening in its strength. You know, the— but, you know, the actual final license, even all the license applications were actually approved for Vogtle and Summerhall after Fukushima. So after that happens, it was actually approved under the Obama administration. Same thing with all the loan guarantees. Necessary to actually complete the construction at Vogtle. Right. So billions of dollars of money is let out by the Obama administration to actually complete— to start construction for nuclear. And then Trump actually continues and gives more loan application— loan money for Vogtle to continue. So we're starting to begin to see— and I think climate change is part of this— that there are people on the climate left, for lack of a better term, right, who are realizing nuclear power has an important decarbonization conversation role to play, obviously, duh, and are beginning to sort of make this a more bipartisan energy source than it ever had been before. And we come to this point now where it's, I think, one of the few things that Americans of right and left actually agree on.

Amazing, we've done it. Uh, and so, so if anyone's Googling at home, Vogtle is V-O-G-T-L-E, and it's what, it's like a city in Georgia?

No, no, it's Alvin W. Vogtle. Vogel, who was the, I believe, a former CEO or president of Georgia Power Company.

Okay.

You don't know who Alvin Vogel is?

No. And so, okay, so what happens there? I mean, like, let's break this down like a case study because, because, yeah, it's not—

I mean, there's a— this, you know, I've done 3 quite boring podcast episodes on this if you want to, if you want to get the, the real, the real depth of this. How do you want to start? So Vogel 3 and 4 are a new pressurized water reactor design called the AP1000. And maybe it might make sense before we dive into Vogel. I don't know. I know you want to talk about how a pressurized water reactor—

Yeah, let's do it.

Versus a boiling water reactor works versus sodium fast breeder reactors.

Yeah. Yeah. So we're— so this is going to be a pressurized water reactor. Yeah. Let's explain it.

Yeah. So, I mean, I think the important thing is that a nuclear power plant is a thermal power plant. What does that mean? It means it takes heat energy and converts it into mechanical work, and that spins a generator, which makes electricity. And the nuclear fission reaction, what it's doing, all it's doing, I mean, that's saying a lot there, is creating heat. And it's sort of like, in fact, the industry thought of it a lot as like a, instead of having a coal coal boiler, you know, furnace, or an oil boiler, oil-fired boiler. We're going to have a nuclear-fired boiler, essentially. So the power plant has to figure out a way of safely generating and sustaining a nuclear chain reaction to generate a huge amount of heat to basically, you know, turn that into steam that expands and then spins a turbine to electricity in a pressurized water reactor like at Vogtle. The way it works is that you have two circuits of water, if you will. You have a circuit of water that runs through the reactor, which we call a primary loop. That's highly, highly pressurized water. And that water is never allowed to boil, even though it goes to 600°F and higher because it's under so much pressure. And then that water goes through these devices called steam generators. Which are like boilers, but they're water-fired boilers. They have thousands of tubes in it that take that highly pressurized water that has passed over the nuclear fuel and passes it through these little tubes. And there's less pressurized water on the other side of it, and that transfers the heat through the tube such that the water boils and creates steam that then goes to a turbine, expands in the turbine, spins this device at 1,800 RPM with millions of horsepower to basically run— spin generator that makes electricity that powers your home and business.

Okay.

And that's what a PWR is.

So that's, that's the AP1000.

That's the AP1000. That's more than two-thirds of the operating nuclear power plants in the United States are pressurized water reactors.

Okay.

The AP1000 is a type of pressurized water reactor. Then we have boiling water reactors, which say, fuck this. Sorry, excuse me.

Oh no, you can swear.

Fuck this. Fuck this, you know, one loop, two loop thing. Let's just boil the water in the reactor and call it a day. I mean, making that pretty simple. So we literally have in the boiling water reactor inside the reactor core, we have boiling water that makes steam that goes out of the reactor pressure vessel, that spins the turbine, that spins the generator, that gets condensed and returned as water.

That— I mean, that sounds less controlled than what you described before.

It's actually in some ways, for complicated nuclear engineering reasons, it can be sometimes more controlled because boiling itself slows down a nuclear reaction, at least in a light water reactor. Right. Because it means just less moderator. Generator, which we can explain what that is if you're interested. But that's, that's the other type of reactor there, and that's about a third of the reactors that are operating in the US.

Okay.

All of the reactors that are operating.

Okay.

A lot of the startups today are focusing on these next-gen reactors, right? So these are reactors that use, rather than water at all, they use liquid sodium, uh, in the reactor to take the heat away from the nuclear fuel, and they transfer it over to a water— to a steam generator with sodium on one side and water on the other. It's an interesting idea.

Yeah. Like, why is that popular all of a sudden?

Well, there's some ideas that, for example, the sodium doesn't have to be under, under pressure because it's, it's liquid sodium, right? It doesn't— it boils at a very high boiling point, so it doesn't need to be pressurized to get to that— excuse me— that high temperature. There's other fuel reasons why these reactors can be better at what's called breeding heating nuclear fuel. Um, uh, there's obviously disadvantages too of, of having liquid sodium next to water. You can imagine why that's a bad idea if you've ever been to high school chemistry where, you know, water and sodium kind of go kaboom. Um, the— there's other reactor types like that, you know, companies like Valor and X-Energy are doing, which say we're going to use helium gas to transfer the heat out of the reactor core rather than using water and not change the phase of the water, just literally move helium through the core over and over again and then run that through a steam generator. So and there's— and then there's molten salt reactors, which literally say instead of even having solid fuel, we're to put the fuel into solution literally with a molten salt and actually have the fuel kind of fission inside the inside the coolant itself and transfer the heat directly from there onwards.

Okay.

So there's all sorts of different permutations. But at the end of the day, what we're just doing is a heat engine.

Okay.

It is just we are generating a huge amount of heat through the nuclear fission reaction and finding some way to convert that into mechanical work to spin a generator to make power.

And I actually, I want to come back to the technology in a second. The— okay. What is the What's the name of the, that 3-part podcast? I forget the guy's name.

Decouple.

Decouple. Okay.

Yeah.

So, so for anyone who wants more information on what we're about to do quite quickly, the Decouple podcast, I did, I listened to all of it. It's fascinating. This, this Vogel story is, is bananas, but let's, let's do the, the 3-part, the quick version, which is, you know, at least partly cost overruns took a long time. People a bit scarred by this. Yeah.

I mean, I mean, I think that the cost overruns, just to be honest,, you know, were more than double. The plant ended up costing more than $30 billion to build 2 gigawatts. It was probably, it was supposed to probably finish in around 2017, 2018. It finished in 2023 and 2024. The interesting part in my opinion is that's not, obviously that's not good. Why did it happen? And I think that the 3 main reasons are the design wasn't done when they started construction. The supply chain that was supplying the plant wasn't actually working when they started construction. And for the specific reasons of the AP1000 design, which is a modular nuclear power plant— a modular nuclear power plant is a great thing, but when you're building modules, like, it's like trying to build a Lego thing. But if your shipment of Legos is delayed made. Well, it's going to be pretty hard to build a Lego, you know, Legoland if you can't have the fucking Legos. And then the third thing really was is that, look, it had been 30 years roughly since we had started nuclear construction in the United States, and the workforce's capability of figuring this all out was, was pretty low. It just hadn't had that experience. That being said, you know, when we tried to build the AP1000 in China and the Chinese had been building a lot of reactors sectors for a long time with a very experienced workforce, it was just as slow basically as it was to build it in, in Georgia, as it was to build it at Haiyang and Sanmen. So I really do think the first two factors were the primary drivers of why that project went so over budget and so, so slow to finish construction.

I might be conflating things because this might be Vogel or it might, might not be. But another thing I remember you talking about that just struck me, which I guess I guess should have been obvious in some ways, but it was, it's that the, the energy company backing some of these projects is, it's basically like an existential project when you, these things are so expensive. They had the math, it has to come online at the right time. And like if you get it done, you're, it's great, but you might go bankrupt on your, your journey.

Yeah. The big problem with nuclear power plants in the United States has been if you, you know, utilities are generally pretty boring businesses, right? Why? Because like even in the worst economic crisis, you know, people still use power. Like, you know, you still cook dinner, you still, I don't know, watch TV, I don't know, use air conditioning. The problem is that a nuclear power plant and the way it's financed in the United States can sometimes be such a large capital investment that you're kind of betting the company that it doesn't work. And the exception where utilities become less boring and maybe go bankrupt is the exception when they try to build new nuclear power plants, right? You know, Washington Public Power Supply System, WPS, as it's aptly named, uh, went bankrupt. This, the largest municipal bond default because they tried to build 5 reactors out of sight, uh, at 3 sites, really 2, but, um, and only finished one of them. And that caused a huge bond default. Public Service of New Hampshire went bankrupt trying to build Seabrook. Um, Long Island Lighting Company was destroyed by trying to build Shoreham and the effective anti-nuclear activists there. Um, and even in case, you know, the Santee Cooper and SCANA, you know, people literally went to jail who were the CEOs of those companies for essentially lying to the Public Service Commission about what was going on at VC Summer. So building a nuclear power plant is a really serious commitment that a power— that a power company is going to do. And it is to immense credit of Southern Company that they stuck through it, even through all these cost overruns, and got those two units online.

And when did they come online?

2023 and 2024.

And that's so that, that's kind of like our crowning achievement over the last 20, 30 years.

Yeah. And, and you know what, we've been bringing some nuclear power plants online, like Watts Bar 2 started online in 2016, you know, started in 1970-something in construction. Before that, Watts Bar 1 started in 1996. Before that, you had some of the last plants of the, in the early '90s, uh, C- Comanche Peak, South Texas Project, Vogtle 1 and 2. Ironically, you know, they were starting to come online just after so many delays. But yes, those were— those are the crowning achievements. And, and Summer did not come online.

Okay. So like, let's get into what you're doing and the current state of affairs. If you're in my chair, I am getting pitched by nuclear startups left and right. And, you know, I'm getting pitched on things called small modular reactors. I'm also getting pitched on, on fusion startups. I'm getting pitched on things like Triso Fuel, which we've done an episode on. Like, you know, it's— I'm just being bombarded with new ideas. I'm seeing in the news Trump is greenlighting, go make these things, let's do this. And, you know, everything that we just talked about for the last like hour is pretty fucking bleak.

So like, it's pretty fucking bleak. Um, well, let me give you— let me just, let me pull back and say, remember Man has been using fire for, for purposes for, I don't know how long, maybe, uh, I'm not an anthropologist, but hundreds of thousands of years, let's say. Right. You know, stole it from Prometheus, right? It has only been since December 2nd, 1942, literally, that Enrico Fermi figured out a way to get the strongest force in the universe and tame it for man. And the idea that it, you know, we're gonna have some challenges in figuring out how to make this totally new fire, which is essentially what nuclear fission is, that has never before been made, you know, controlled, manipulated by man into a source of effective energy. That's not surprising that, that it was. What's in fact amazing, as you say, it's bleak, but at the end of the day, 20% of the power that this country makes is made from something that had not been discovered until, you know, 1942. And that in France, 70 to 80% of the power is made from this technology. So it is, um, yes, we've had challenges. I, I, I— you can't argue with that. But the fact is, is that a lot— lights are like— lights would not be on in this country if we hadn't figured out how nuclear power works.

Yeah, no, that's It's good to have these.

It's really important that we recognize that this is a totally new— it is something in the history of man that 1,000 years from now, people will recognize that moment when we got access to something that was incredibly powerful, thousands of times more powerful than the chemical reactions that power combustion. So it is like, yes, I totally agree. Like, it should have gone better. It could have gone better. France maybe is an example of where it did go better. China probably will go better, but like, it's also, we're, we're, we're working on something that is like a, something that a god gave to us, you know, in Chicago in December of '42. And it's gonna take us a little bit of time to figure out how to, how to properly harness it and use it to power the world.

No, no, this is, this is a good perspective. So let me, all right, so William, can I, can I set up SMRs for people just for a second? At least, at least in the context of some of the stuff we talked about.

Yeah. I wanna get to my company, but like, yeah.

No, no, we, okay. No, we're gonna, we're gonna, we're We're going to dig deep. The, you know, so you were just going over the existential bets these energy companies have to make. I mean, part of the pitch that I get for SMRs is that it's small, so you don't have to go all in. We're just going to build these as we need them. And then people were already working on this, and then the AI boom happened, and it's like, okay, and now we need, we need more. Let's just go put these things right by the data center. Sure. So I mean, there's like 8 SMR companies, maybe more. Oh, the—

depends on how you define. I think there's probably, yeah. 100 at this point.

Yeah. So, so that's— I mean, is there a way for us to— it's hard to summarize.

So let's give, let's give the advantage right ahead of the SMRs right there. It's that at least on paper, and we should just do it between a microreactor and an SMR.

We should define— I kind of went quick. We should define for people.

Yeah. So like, you know, the average nuclear power plant is 1,000 megawatts and change. Right. And there are people who are starting in— well, this has been going on for a while— have basically said, well, well, part of the reason why you're having to make such a big bet on this is even if it's actually pretty cheap per kilowatt, when you have 1,000 megawatts, it's a lot of kilowatts, right? It's a million kilowatts actually. And so therefore, just the price is going to be— the price tag is just huge. So you're betting your company to figure out how to finance to buy this thing. And so wouldn't it be smaller? Rather, let's make it instead of 1,000 megawatts, let's make it 50 megawatts or let's make it 200 megawatts. Megawatts, great. And then there's some people who say, fuck even that, let's make it even tinier. Let's make it 10 megawatts or 5 megawatts or 1 megawatt. And let's make just it all in a factory, print it out, we get mass production, get down a cost curve like a car, right? Like, you know, like Henry Ford did with the automobile.

Instead of building a supercomputer, it's a cluster of chips.

Yeah, exactly. And it will just build 1,000 1-megawatt reactors rather than building, you know, one 1,000-megawatt reactor. Why am I not optimistic about this approach?

Two reasons.

We can go in the history and then the science. So the history is the nuclear industry started, of course, on very small reactors, right? If you think about, you know, obviously the— that reactor that was built in, in that squash court in Chicago was about 200 watts. So let's call it a nano reactor. I don't know, or a pico-femto reactor or something. But I think more importantly, that wasn't a power generating reactor, in all fairness. So, you know, the first reactor experiments that we were basically building in the early '40s, late '40s and early '50s were really small and they actually weren't light water reactors at all. They were exotic using these liquid metal coolants. Like the first real reactor that we built a sizable amount of power was something called Experimental Breeder Reactor 1, which was, you know, a sodium potassium eutectic cooled fast reactor that was very small, could probably fit in the size of, you know, the Playground office, probably even smaller than that. And I bet you I could probably get some seed money. I could go on, you know, some VCs right around here maybe and get some funding to build that exact reactor and say there's all sorts of reasons why it will work. Work better. And then the first actual real commercial nuclear power plant, which was a shipping port, was 68 megawatts electrical, right? So right in that SMR space that was designed actually using a lot of these principles, right? Like, let's try to make this using off-the-shelf components to basically make a small plant like an SMR would be today. Why did we suddenly go? Why did the industry scale by an order of magnitude? You know, Shippingport comes online in '56. By 1968, we were building 1,000 megawatt reactors. We scaled by order, an order of magnitude more than that over less than a decade. What happened? It's that like all power, thermal power conversion systems, right? A nuclear power plant is immensely impacted by economies of scale. And You can look around your daily life, look like go out to that parking lot, you know, 75 feet from where we're standing right now and see what's going on. You know, and when we think about mass production, we think about the automobile industry. Literally Henry Ford invented, you know, at least according to Henry Ford, the mass production system, right? To basically make automobiles. And if you think about a company like GM today, General Motors, not even one of the largest, automobile companies, they're making 100 gigawatts worth of internal combustion engines every year. But no one in the power generation sector basically says the right way to power America, even if we want to use fossil fuels, is stacking 1,000 or 5,000 Chevy Tahoe engines together, even though those are totally mass-produced, have an extraordinarily efficient just-in-time global supply chain that is getting costs really, really down. Why don't they why do that, especially when it's already been done? The reason is it just is a lot cheaper and a lot better to build a 100-megawatt or 500-megawatt gas turbine or oil boiler than it is to basically do thousands and thousands of smaller fossil fuel engines. And there's some real intrinsic thermodynamic and geometric reasons. Like, I don't want to go into the, you know, square-cube law, but there literally are reasons that you can figure out from thermodynamic first principles principles why it makes sense to go much, much larger. And then with nuclear reactors in particular, those— all those thermodynamic constraints comply, but then all the issues with shielding all of a sudden become even more— make this even more profound. You know, if you think about a 1-megawatt reactor, if you were standing next to a 1-megawatt reactor that's on, you get a fatal dose in milliseconds, right? The The difference— and so we have to build shielding around it. And the shielding distance that you have to build, the concrete, the steel, they have to build just to prevent those death rays that are emanating, those gamma rays that are emanating out of the plant, from out of the reactor, from doing— they are basically, you know, the same to, to the first order, right? Uh, whether it's 1,000 megawatts or 1 megawatt, right? And obviously there's a little bit of difference here, but not very much in terms of that. So we have all the thermodynamic reasons why it makes sense to go bigger on fossil plants that also apply to nuclear plants. And then all of a sudden you're adding all the radiation shielding and safety system issues that you need to make— operate a nuclear power plant safely, which is exactly why the industry went from those smaller reactors and scaled them up. And every time that we've looked at any country's nuclear industry, they always are starting out smaller and they always start scaling much larger for these reasons. So you can't get around physics. It doesn't matter what regulatory system you are in, the physics are—

the physics.

China is doing this. When the French and British advanced reactor program was doing this, when the Russians, which have the most advanced sort of non-light water reactor program of anyone in the world, right, they always are scaling their plants larger to try to get these economies of scale, to get the unit costs down.

It's like rockets. Everyone starts small and ends up at the biggest possible rocket they can make.

Yeah, sure. I mean, like, even, even— I mean, I see, like, I'm imagining that's— is that Starship right there?

No, that's a Falcon 9.

Yeah. So if you think about Falcon 9, right, the Merlin engine, right, gets replaced by the Raptor engine. The Raptor engine is a lot larger of an engine. Those economies of scale are even more profound in a nuclear thermal conversion. Yeah, power conversion system.

So then why are— I mean, if— why are people trying this?

Well, I think people are trying this because they've looked at the challenges of the large light water reactors and said that's not working, obviously. And like, we need to figure out some other approach around this. I just think that we're kind of re— you know, we're fighting a battle that can't necessarily be done. And then to steelman my argument, what they would way is, look, if we can actually get this mass production of these components down, right, like, yes, we'll lose some economies of scale, but we'll make more than up for that from the mass production. And what I say is, look at the fossil energy industry. There's a reason why we don't build most of the power that comes from the grid. It's not built from 5-megawatt or, you know, 750-kilowatt, uh, um, you know, reciprocating or, or, or combustion turbines. Like, those economies economies of scale are really obvious, even if we have a highly mature power conversion system industry like the internal combustion engine industry that doesn't work. And that's, that's why the industry itself, when we were building tons of reactors, scaled so large so quickly to capture those economies of scale. And I don't see any reason why we, why we can change those physics anytime soon.

And so all this money is chasing these SMR startups. I mean, billions and billions have been raised. Sure. And, and, but I mean, so you think fundamentally flawed?

Well, I mean, I'm raising tens of millions of dollars, maybe from even inside this building. Yeah, I have raised a lot of money based on that, the idea that premise is incorrect, that the same physics that applied in 1958 and 1968 and 1978 and 2005 applies today, right? That electric power is a commodity that if we're gonna, you know, access the electric power industry, which in the US, just to give you an idea of how big this industry is, it's a $500 billion plus industry. It's bigger than the entire advertising industry is just what electric power, the gross electric power sales is just in the US is. So we're talking about a motherfucking massive industry that is in need of power right now. And we have the biggest companies in the world, maybe some of them have headquartered 5 miles from where we stand today saying, I have the biggest balance sheet in the world and I need power at whatever price or at a price that you can do it. And anyone who can beat that price, beat that timeline, they're gonna make a lot of money. And that's the basis of the company that we're founding, we're building.

Yeah, so let's, okay, let's talk about, so the, I mean, you've explained it to me before, but obviously everyone wasn't there for that. Yeah, no.

I hope you weren't recording that.

So Alva Energy, you guys are based in Massachusetts.

Yeah, Cambridge.

In Cambridge. And yeah, I mean, so you have this, you have this very smart, more efficient kind of direct shot to getting more juice out of our existing.

So we have, we have a different thesis about what went wrong in the nuclear industry. Right. Let me, let me set us up there and then tell us what we're doing. So our thesis is, you know, we actually did build a lot of nuclear power plants in the United States very quickly. Very quickly and very cheaply. They weren't the majority of them. The, you know, the majority of them weren't bad. But there are these exceptions— St. Lucie to Palo Verde, etc.— that went really, really well. And we're like, why did those go well and other ones went bad? And what we came to was, is that part of it had to do with the structure of the U.S. utility industry, that U.S. utility industry right now, although they were historically responsible for building the power plants, right? That's not their main job anymore. You know, it was the job in the '50s and '60s when they just had to build so many power plants to keep the lights on. But even now we're at 2%, 3% electric power growth per year. That's not what they're doing. Most of what they do is operate. So we said is the best power plant companies, the best power nuclear power plants were built by utility teams that had a huge development organization inside of it, and they were able to figure out how do we actually build these plants cheaply and quickly. And we said, how do we actually do that? How do we rethink the Edisonian utility model and say, let's actually figure out how to systematize that and standardize that? And then we figured out that inside the US nuclear power plant fleet that is operating today, there is between 6 to 10 gigawatts electrical of additional capacity that we can unlock with some quick, standardized modifications.

Per plant?

No, no, no, in the whole thing.

Okay.

Okay. So, so that's just to give you an idea about what all of Google, all of Microsoft, and all of Amazon domestically use combined. We can just get out by modifying the existing, uh, plants.

Wait, let's do those numbers just so everyone's on the same page.

All those, the large data center companies combined, 3, the 3 of them in the US.

Yeah.

Use around the equivalent of 80 terawatt hours of power.

Okay.

And you're saying we could, we could to find that plant in the existing plant. Okay. Right.

Okay.

And so what we have done is developed a new technology that allows us to bypass— and this process is called uprating. It's taking this plant that was made, designed in 1970 and figuring out through better computer simulation, through new steam generators, right, through modification to the plants, a way to get a lot more power out of that existing power plant. And then we built our own proprietary technology on the balance of plant on the turbine side to actually bypass the primary driver, the primary thing that slows down power upgrades, which is on the steam turbine side. Right. So if you think about a nuclear power plant, as I said, you have the nuclear steam supply system that makes steam and you have a big turbine plant that takes that steam in a pressurized water reactor that's non-radioactive and converts it into electricity. If you're going to upgrade these plants by 20 to 30%, which is what we're doing, right, to actually upgrade that massive 1.2 gigawatt steam centerline, that turbine, that feedwater heater, the condensers, the generator. That takes months and months and months, and you're kind of jury-rigging a 55-year-old machine to make 30% more power, which is sometimes very difficult, if not impossible. And the plant needs to be offline for that entire process. So we said, we're not gonna do that. We literally built instead a, what we call a second turbine generator plant, which sits along the, on the side of the plant and literally takes steam upstream of the main turbine. It takes all that uprated steam, diverts it, can be built while the plant's puts online, generates it, and leaves the original turbine generator plant unmodified as if the upgrade never occurred.

So there, there's like, there's steam that is just not being— all the steam's not being—

it's not that there's steam not being used, that the reactor— we can modify the reactor pretty quickly to actually make it make more steam. The problem is we don't sell steam in a nuclear power plant, we sell electricity. And actually figuring out a way to modify the turbin plant to actually take that more steam and convert it into electricity is actually where the cost and schedule risks of these power upgrades lie. We have a totally novel technology that says don't modify that plant at all. Just literally build a second one while the plant's online in a standardized modular way. It's a 300-megawatt steam turbine. Take that excess steam and basically, um, hook it in when the plant goes down for its regular refueling. Refueling outage, bring the reactor up to the higher power, and then take that excess steam that now that higher power is generating and put it over to our second turbine plant.

And this, this regular refueling, um, yeah, how long?

Every 18 to 24 months it occurs, and they take around 31 days. We can swap out steam generators. It's been done many times if the plants, you know, Hatch can take it in about 54 days. So instead of being a 31-day outage, it's 54-day outage.

Okay. Okay. So when you were explaining this to me the first time and I, you threw a lot of information at me after we talked, I mean, this was my biggest question is like, I get what you're saying, but there are ifs, right? It's like if we can hit that window.

What do you mean if we can hit that window?

Well, so you're saying we know, we, we know we can do this, it's been done before, but I mean, these, these companies would be taking a huge risk that you're not gonna fuck things up in that window.

That's a great, a great question.

You know, and, and that you're going to be in that? Because if it's— if you don't hit 51 days and it takes a year or something to do this, then—

So what's important, I think, to recognize is that the thing that we're doing on the nuclear side and the pressurized water reactor side is we're not— we're modifying, we're replacing the steam generators. The steam generators that we're modifying with them are not new steam generators. They have 110 effective full power years of experience in the global fleet. So we We know they work. They're not going to break the plant that way. And we've replaced dozens— dozens of power plants in the US have already replaced their steam generators. Right. In fact, the president, the SVP of projects for my company, set the US record for the fastest steam generator replacement ever done, the second fastest in the world. Beat it by 4— lost it by 4 hours for the world record. Right. So this has been done a lot in the US nuclear industry. This is not anything new. And we know those steam generators work. And we know that this general upgrade approach that— excuse me, replacing those steam generators can enable this large of a power, power upgrade because it's been done before. I was just in Sweden last week at a nuclear power plant called Ringalls, and Ringalls Unit Number 4 upgraded their power plant by 21% using this exact approach on the nuclear side of the house. What was slow at Ringalls was figuring out how did we take that existing turbine island and make it, modify it, and jury-rig it to take 21% more power. So in other words, everything on the nuclear side of the house has been done before, is using components that have been proven in the fleet, and are using approaches that are totally standardized and have proven themselves in the real world before. What our technical niche is, is saying, well, the real reason, if you talk to utility right now, is they say, okay, this sounds great. I can make the reactor plant make 20 to 30% more power. Fantastic. The problem is I don't want to have my turbine open, splayed open, my generator splayed open, my feedwater heater splayed open. And that might take 9 months, that might take a year. So how do we get around that problem? We literally, we literally connect into the steam header with a simple sort of, you know, it's a saddle if you want to know exactly how it's done. And same thing with a feedwater header, same with our INC. And let's Imagine this doesn't work. Now, this is what we always say to the utilities, it will work. But imagine that second turbine plant doesn't work for whatever reason. Our instrumentation control solution doesn't work, whatever. What does the plant do? The plant literally, sorry, literally just shuts that steam valve, shuts that feedwater valve, and operates the plant at the pre-uprated power with new steam generators that are proven. So the risk to the plant is extremely low because of the solution that we have devised. It can literally separate and segregate from the existing plant if it didn't work or if there's any operational challenges. So it's a really low-risk modification to get an SMR size worth of power between 200 to 300 megawatts per reactor for 1/7 to 1/10 the cost of an SMR.

And you're saying we know this will work, we've seen in the past, but you guys have— you're a young company, you have not done this yet. When's the first time you're going to try?

I mean, that's under NDA with multiple utilities right now. But let's just say we think that there's a pathway for us bringing power on using this approach within 3 years.

Within 3 years. And I mean, you mentioned it briefly before, but I mean, okay, I don't know how to say this artfully. You know, there's—

Say it inartfully.

You know, there's a— it's Silicon Valley. It's like you We want to be an SMR. We're doing something that sounds new and it's flashy. And then you guys come along and you're like, hey, hey, no, we have a simpler thing. We're just going to kind of retrofit. There's a little less sexy, you know, of the sales pitch, even though it makes economic sense and everything.

But this is the first stage of our master plan. I mean, do you want to talk about sexiness here or the lack thereof?

It's just funny because you guys, what you're doing sounds so reasonable, rational, practical. And then, you know, but in Silicon Valley, in tech, you get rewarded for the more off the wall, you know, One of the—

when we were— we, we did this whole project where we studied, um, a bunch of, uh, uh, we studied every nuclear power plant ever built in the United States, and then we studied about 2 dozen around the world, and we're, we're putting the company together. And we went out to Austin, Texas. Uh, one of our investors, Isabelle Bemke, uh, her husband's Joe Gebbia, who's on the Tesla board, co-founder of Airbnb. And we were meeting with a a Tesla guy and a SpaceX guy. And, you know, one of the lessons— I won't name them— one of the lessons they always taught, taught us, and I believe this, is scaling strategy is everything. For hard tech, scaling is everything. If you think about— not to, not to sing Elon's praises too much, but if you think— and what a cliché— but like, to— if you think about SpaceX and you think about Tesla, you know, Tesla started with the Roadster. SpaceX started not with the Falcon 9, but with the Falcon 1, right? And then they led to the Falcon 9, which led to Starship. Tesla, same thing. The Roadster was essentially a Lotus Elise battery, a Lotus Elise frame with some laptop batteries and some electric motors in it. But that was the foundation of the Model S. We had the exact same approach. We don't think that the correct approach here is to immediately start building new reactors right away. The foundation of our company is predicated on the idea that we have a— we have figured out a rather unique scaling strategy for our company, which are these power upgrades, and that we can deliver, build our organization to deliver nuclear power faster and cheaper than any other technology at a very, very high capacity factor and deliver gigawatts of new power to the US economy in a time when it desperately needs it faster and cheaper than anyone else. But what we're doing right there is building that exact same development organization organization, that exact same skills, technology, and virtual digital construction in schedule control, in our supply chain management, in our regulatory interface and affairs, right? That's actually going to be proven in the real world and use that as a springboard. Once we're done with a phase 2 of the Alva master plan, is literally after we are done with these upgrades, we will have— be the best organization in the world to build new reactors. And that's exactly the strategy that actually proved— that actually powered successful nuclear power plant builds that happened in the 1970s and '80s. The problem was, under the old way that the utility industry was structured, is that that team would be locked up in a utility. And once that utility was done building their reactors, they literally never built another nuclear power plant ever again. Because we are utility agnostic, we can take that exact same tech and capability that we're building on the power upgrades, deploy it right along to the power— to the new builds. Builds, and we can do it anywhere in the world. And that will be the first time that structure and that tech is deployed in that way. So the power upgrades are our scaling strategy. It just happens to be a highly lucrative and highly, um, uh, important scaling strategy. It can deliver a lot of power very, very rapidly.

And what's the piece that you guys have invented? I mean, because some of this seems like you just figured out, um, a smart thing to do that The second turbine. Okay.

That has never been done on any U.S., any nuclear power plant in the world.

But you're not building the turbine.

But we're building the instrumentation and control that interfaces it.

Okay. Okay. And so, and that is because I know nothing about, you know, what that entails. I mean, so in that part is, is totally proprietary, patent protected, etc., etc. So you buy the turbine.

Yeah. And you get—

there's lots of turbine manufacturers and then you hook up your control system to it. And it's this control system that has, I assume, software and sensors and all these things that are kind of proprietary. Indeed. And, and just on the turbine piece, I mean, you just said they're available. Are we in like some huge turbine shortage?

Great, great question. Thank you. I wish I had given you that question. Had I prepped better, I would have emailed, ask me about the turbines. So there is a gas turbine supply chain shortage. Okay. What's the difference? So a gas turbine is like a jet engine, right? Basically, it's what we call a Brayton cycle in thermodynamics. It basically takes a big combustor, burns gas, and then basically uses air, compresses that air and expands that air to spin something. The steam turbines, which are what we are doing, do not have a very large supply chain backlog. Without violating any nondisclosure agreements, we have quotes from name brand steam turbine manufacturers that are quoting delivery of a 300-megawatt turbine in 24 months or less. So there is a gas turbine supply chain shortage. There is not a steam turbine supply chain shortage.

Okay. And if these companies, these energy companies could get this gain and it's just been sitting here all along, like, why has somebody not tried to do this before?

There's two reasons. One is that it's important to recognize that just 5 years ago, they— we were shutting down nuclear power plants that were perfectly safe, perfectly technically, you know, from an engineering perspective, perfect. We were shutting them down because they couldn't compete with the power market that we had at that time. So if you're thinking about a power— if you're thinking about a utility at that time or even a technology innovator at that time, they might be able to figure this out. Maybe it took a— it was a— it wasn't so easy to actually figure all these pieces and put them together, but there wasn't the market. Literally, the idea of spending hundreds of millions of dollars to increase the output of the plant, which when you can't basically make that plant be economically feasible today doesn't make very much sense. And what happened over the last 5 years is that all of a sudden we are in the, for the first time, nuclear energy is being paid at a very large market premium compared to everyone else, right? You know, we saw Microsoft basically buy power from Constellation at roughly double to triple what the wholesale clearing price of electricity is at in PJM normally. They are— these companies are willing to pay massive premiums to do that. And while the rest of all the startup and venture world were sort of, you know, saying we're going to build, you know, sodium-cooled, you know, water-moderated, heavy water moderated reactor or whatever, we said, what is the fastest way that we can get power to actually seize this grid and build and scale our company to build large light water reactors as fast as possible? And it took us— I'll just be honest, it was a year to figure out how to do that. This from a thermal hydraulic and nuclear engineering perspective.

And you think it'll take 3 years before your technology could come online?

We are— we are— we have active and very serious discussions with 5 different utilities right now.

And this is a US thing?

Oh, we're only focusing on the US market.

Okay, okay. I just got back from Nevada, um, like couple days. I— the amount of data centers going up that I saw in Reno is obscene. I mean, and God, yes. I mean, I— and obviously I'm covering AI all the time. I I mean, these people want— they want some— they want some power. They want—

they want power as fast as possible. And I think that what we're trying to say is, okay, let's give them gigawatts of power now, and let's build a scalable, sustainable organization so that we can give them terawatts of power in 10 years.

And your master plan, I mean, it's— you're saying you would go through this process a bunch of times, and you have this turbine—

can you have this turbine? We're doing some— you're replacing the steam generators, we're doing all the regulatory work necessary to do that. We're sort of, if you will, getting the muscle, the execution muscle, while actually delivering power to then go ahead and actually build new reactors afterwards.

Just because you would have gone through this process of working with reactors for so many years and so many times—

not so many years. Well, but I mean, I mean, a number, you know, and financing and, and all that. That is the thing that has proven itself to actually be the critical determinant of what actually makes a cheap versus expensive nuclear power plant.

But how would you gain the institutional institutional knowledge of how to build the plant in the first place?

So the institutional knowledge of building the plant is much less of like the actual, let's say, reactor technology, right? We're going to be building probably a— well, I'm not going to tell you which reactor we're going to be building, but that reactor has been built before. It is actually the execution muscle. It is how do you actually keep track and package your work packages properly? How do you manage your labor supply? How do you manage the Nuclear Regulatory Commission? How do you build technology to do all of that work? That is not a trivial task. When you think about how much a nuclear power plant is. This— we're talking about over 100,000 work packages just to finish a nuclear plant.

What's a work package?

Yeah, that's a great question, right? This gives you an idea of how kind of complex this, this entity, these entities are, these power plants are. A work package is literally the package that goes out to your, your welders, right, your civil engineering crew, to basically say this is what you need to complete over this interval to get the project to the next step.

Okay.

So it has all the designs in it, it has all the regulatory requirements, it has all the supplies that are needed, it has what the weld traveler, welding travelers, what the QA processes are going to be. That is where projects go right or where they go wrong, as an example. And when we say when a plant is design complete, what does that actually mean? It means that the, the design is actually buildable in the work package. The work package doesn't require going back to engineering to figure out something else to actually change in order to make it work. In the case of Vogel, there's so many work packages literally that would, were physically impossible to build. Like there would literally be a physical interference happening in the same place, in the, where they're trying to actually build something at the workspace. What we're doing is we're saying, let's actually take that exact sort of execution engine that is necessary, all the work packaging, all the supply chain management to make sure that our products actually and our inputs actually arrive on time, all of our regulatory shop, let's build it on a process that is about 1/10 the capital size of a new reactor build. Let's get all that technology that we're building in-house proven, dogfooded, deployed over and over again, get power to the grid faster than anything else, actually prove to our financing partners, to our regulatory partners, to our utility partners, and our data center partners, we are the counterparty to do, and then take that entire organization, all that build tech that we have developed, and then turn it on to a large light water reactor build.

Okay. Okay. So you just walk through like why people would trust that you're going to figure this out. If you, you know, if you Google you and you— I mean, look, you've just proven your knowledge of this industry. However, you have an unusual background to get here. If I Google you and I go to your LinkedIn page, I'm seeing a guy that has a physics degree and then spent a lot of time in health care, you know, and relatively recently, if you're just looking at the— Yeah, that's great.

It's a great question.

You just appeared on the nuclear scene.

Why? Why? How did I— How did I— So as I think I mentioned, my father was a nuclear engineer.

Yeah.

7 reactors. So I sort of grew up in this, in this industry. And, you know, I spent 10 years doing a lot of work on scaling biomedical interventions to the populations, to populations in the US and around the world. How do we manufacture them faster? During like the COVID—

I did a lot of work on COVID Vaccine manufacturing scale-up.

So one might say—

wait, sorry, real quick. Your dad was a nuclear engineer for his whole career?

No, not for— so the nuclear industry sort of died, right? He graduated in 1979. He worked on Sharon Harris, St. Lucie, South Texas Project, uh, go through Washington Nuclear Project 3, you know, it doesn't matter. And so I spent literally my fifth grade— my fifth birthday party in a control room of a nuclear plant.

This is what I was wondering from your other chats, and nobody got into this. You were And you were into it.

Yeah, I was super, super into it. But to be honest with you, I didn't want to go into it immediately. I wanted to figure out how other industries scale their technology, how other industries work, how the politics— I spent a lot of time working on politics. How do we get public acceptance of new technology, public policy support for new technology? And then I wanted to meet the people in the industry who I wanted to work with. So that's how I met Robbie, right, who has a PhD from MIT in nuclear engineering, who's my other co-founder who spent 7 years at General Electric designing and building jet engines, the most powerful jet engines in the world. How I met Rich Kalman, who has done $20 billion worth of, of nuclear construction in his career, has set multiple world records. How we met Steve Mannin, who is our regulatory engineer, who set the world record for the fastest design certification in the history of the Nuclear Regulatory Commission for the GPR 1400. And thinking about how to actually build a world-class team as a startup took me, I'll be honest with you, 5 years. And thinking about what I was doing in the other, in my other careers helped a lot because I saw how was, what's a really example of an American industry that is highly, highly regulated and highly effective actually producing and surging production? Well, that is actually the biopharma pharmaceutical industry. And I think that sort of work that I did there was very, very instructive into how we're going to be scaling nuclear— my phone just dropped— how we're going to be scaling nuclear plants, uh, today.

And this, this knowledge that you have, I mean, you, you know, when I first met you, and anyone can hear it, I mean, you sound like you're an MIT PhD in nuclear engineering who spent 40 years working on this stuff.

An MIT PhD in nuclear engineering, though, would not be spending the time on building— I mean, one of the problems that we have in nuclear pedagogy today is they would be much better than I would be at MCNP or at RELAP or, you know, TRACE or something like that. They would be working on the thermal hydraulics of a nuclear reactor core. I don't think that's where the problem in the industry lies today. Part of the issue with nuclear, nuclear engineering pedagogy today is because the industry has not built for so long, so much of what actually is going on at MIT's nuclear engineering department, which we have a very close affiliation with. We have multiple graduates of the MIT nuclear engineering department. We literally are 1,000 feet away from them. And this is not a read of them, but they have been focusing so much on what we call paper reactors, right? These are what— if you want to do a PhD dissertation in an industry that's not building, well, what are you going to be doing? You're doing computer modeling of a new reactor type. And that's all great. That's all fantastic. But it doesn't actually, in my opinion, get to the heart of where we've had challenges in this industry, which is actually building and executing. Executing new plants. Right. And that's a product of one of the big downsides of us stopping nuclear construction basically altogether for 30 years is, is that the jobs that were available if you were getting a nuclear engineer was going to a national laboratory or becoming an academic. And there's nothing wrong. We need national labs, we need academics. These are really important parts of the US nuclear energy ecosystem. But that's not where the challenge has been actually in in building new plants, right? It's not— it hasn't been in the space of, you know, that the react— the neutronics have been wrong, right? That we are really good at neutronics and we're really good at making nuclear reactors today. We're really bad at building them.

Okay. Okay. Let's do the James makes some enemies lightning round if you're prepared for that.

Oh, no.

Oh, before we do that, Alva Thomas, Alva Edison.

Yes.

Yes. Okay. I think this is going to be okay.

I think it's going to be okay. I'm nervous already.

No, no. I just want to go through, look, I went out to Tennessee. I went to visit Standard Nuclear. Yeah. And there it's a bunch of people from Oak Ridge National Lab. Kurt Sorensen is a super smart guy.

Triso.

Making this Triso fuel for these small modular reactors. Super smart guy. Nuclear physicist.

Everyone I talked to was— I'm sure he's brilliant.

Yeah. And he's He sat in the same chair you're in, well, sort of, in previous office, previous studio. Probably was the same chair. You know, and made a very good case about why SMRs make sense, why their approach makes sense. Oak Ridge has been working on this stuff for decades. I mean, so when these very, it doesn't have to just be him, but talking about the SMRs in general, I mean, these people are smart, they're rational. I think they believe in what they're doing. Doing. So they are what? They are just, um, they've talked themselves into this. I don't think that—

I think it's a little unfair to them. Um, I think there's two things that are, that are going on here in the U.S. industry. I think one is nuclear engineers like to solve every problem with a nuclear engineering solution, and some of these are not nuclear engineering problems. So take Triso. Triso, actually from an inherent neutronics from a ceramic perspective, has what we call worse burnup than light water reactor.

And it's a fuel.

It's a fuel type. The primary case, as I understand it, for TRISO is it can avoid the large containment building sort of that may or may not drive the cost of a nuclear power plant.

So yeah, can I— I mean, I'll try and go ahead and fix that. I mean, so they— I mean, they make these little— these— they coat the uranium in this like ceramics and graphite and different things to make it safe. And then this containment you're talking about is that if we're going to make these small modular reactors and you want to place them close to buildings and people and things, you don't want to have to have this massive space. So the whole pitch is, you know, blocking it off. And so the pitch is this fuel is safer. We've been working on this for decades and this— and now we need somebody to mass produce this fuel. And there's a ton of small modular reactor startups that want to use TRISO.

Yeah. So TRISO is not new, right? You know, we had reactors in the 1970s using— we've had reactors in Germany using TRISO in the '80s and '90s, and the Chinese, right, have been building TRISO reactors, right, and have had a 30-year development program to build a TRISO fuel high-temperature gas reactors. So they started with HTR-10PM back in the late '90s, operated a 10-megawatt reactor. Obviously, that's not particularly economical. So they just built a, you know, a big 250-megawatt electric electrical reactor called Shidō Bay Unit Number 1. And here's, here's my answer right away. So they have been at this for 30 years, have had an incredibly, incredibly aggressive R&D program that's government-sponsored. Billions of dollars have been spent. They license all the world's high-temperature gas reactor and Shidō technology from everyone else. They have built a— not a, you know, non-critical reactor. Built a reactor, 10-megawatt electrical reactor, and operated for decades, right? All triso fuel, pebble bed reactor. And they built the first commercial reactor, you know, and it turned on 3 years ago, went through 2 years of non-commercial testing, and then entered its first year of commercial operation. What was its capacity factor when it operated? 20%. So that reactor—

and we're at 90%.

So that's less than a solar panel would be at that spot. So that's fine for the Chinese to do, right? And that's because this is a different fuel type. It's a different coolant system. All that sort of learning that was necessary to bring the light water reactor fleet from 55% to 90% over the last 30 years, none of that applies. And that's why when we built a trizone-bizofueled reactor in Colorado, there was 300 megawatts electrical, it operated over its 15 years of commercial operation at 16% capacity factor. The problem that I see here with TRISO reactors is not that TRISO is bad. TRISO is a great technology. I love TRISO. The problem is, is that our ability to make a high temperature gas reactor—

that's going to be the clip.

Yeah, there you go. I love TRISO. I mean, well, there's nothing wrong with TRISO. TRISO is a great thing. It has a lot of applications for a lot of reasons. The question, though, is what is the— we're not in an industry that that is about selling really cool engineering topics, right? We're in an industry that has to be able to compete and sell power at a very low cost, at a very high reliability. It's gonna take a couple of decades probably to get the learnings necessary to make a triso-fueled reactor work. That is something that we should do, but I don't think that's a ventureable sort of company at this stage. And I don't think that it is the fastest way to new nuclear in what an actual customer wants. A customer wants a, you know, source of energy that's operating on 90% capacity factor. And just to end this, there's a reason why the Chinese are building maybe 2 or 3 TRISO-fueled reactors and building dozens of light water reactors, because the technological maturity is there for the LWRs. It's growing on the high temperature gas reactor side, but it's not there yet. And I don't believe that the point of, you know, a ventureable company in this market is to basically run, you know, science experiments.

Okay. Okay. And then this next question, I have to confess, my knowledge on, on Valor is largely pictures on X. So I'm at your, your mercy here. But I see the startup that seems to have been born not too long ago all of a sudden is making small modular reactors. That are flying in planes to Utah to some government thing? Like, please tell me. They get a lot of attention.

They have great Twitter game.

Yeah. Like, what are they doing?

And I think that—

why does it seem they move so much faster, at least cosmetically?

But what does faster mean here?

Okay.

So, like, I mean, that's a genuine question.

I saw— I feel like they've only been around for what, like 2 years or something? Sure. And then now they've got a—

they had some kind of reactor in a plane flying Well, they had, you know, pieces of a reactor that was shipped in an airplane that had no nuclear fuel in it. I mean, we have been able to ship things in airplanes for a while. Thank you, Orville and Curtis Wright. Or not Curtis Wright, Orville and Wilbur. There we go. No, no, in all seriousness, I mean, it is impressive how fast Valor is moving, but I think they are coming to this general point, this problem of how do we get high temperature gas reactors to operate at a high capacity factor, period. Like, that is what the money is in this industry. The industry is what— if you talk to Google, Microsoft, Amazon, etc., they will say what we are looking for is power that is 90% capacity factor plus. And getting the technology to figure out how to actually do that, you know, it was challenging enough to build light water reactors. But when we, we did have a high temperature gas reactor industry in this country, right? General Atomics built Peach Bottom One and it built Fort St. Vrain. And then in the Brit— the Brits, you know, we talk about the French, right? The British actually made the exact same decision as the French did, and they made a very fateful decision. The British said, we're not going to go with light water reactor technology, we're not going to do what the French are doing, which is basically copy these pressurized water reactors. We're going to build a high temperature gas reactor program called Advanced Gas Reactor program, and that they, after building a dozen of those reactors, dozen plus, they still can't compete with the one light water reactor they have in the United Kingdom. The ability for us to get these reactor technologies to a point that they are highly reliable and able to generate power economically is an extremely, extremely difficult challenge that has bedeviled the global nuclear industry, um, over and over again. And my basic question to Valor— it's an actual question, it's not like— is how are you going to basically make a high-temperature gas reactor that is economical and can operate at a high capacity factor? What, what, like, without having the debugging time that is intrinsic to this sort of power generation technology, right? And I have no doubt, by the way, that Valor can go critical, that they can make a fission reaction happen. The K effective can go above 1. The K effective for most of these, like, that is, that is once again physics that we have known since 1942 before there was electronic computers. We've been able to figure out how to assemble fissile mass in the correct geometry to make a self-sustaining nuclear chain reaction. That is not the hard part. The hard part is how do we take that heat heat source and convert it into usable work that is highly reliable. It's the difference between a campfire and a steam engine. You know, building a campfire, you know, a lot of people can do. Machining and building a steam gen— like a, a, a, a steam engine that makes useful work, that's pretty difficult. And figuring out how to make one that can run constantly and run maybe 18 to 20 months continuously with a very, very high reliability. That has taken the global power industry decades of technological experience to actually get done, uh, billions of dollars of development. And I, I've yet— and maybe I'm missing something, and I'm open, um, to, to being wrong here— I don't see that scaling plan. I don't think just because you get the new, you know, K effective above 1, you get the self-sustaining chain action running, that's like square one of 10,000 steps to make this into an effective power plant.

Okay. Okay. So that's for Isaiah.

Can, uh, come on the next episode. Yes.

Yes.

But on the other hand, I mean, let's just be frank. It's also very impress— I mean, kudos to Isaiah. I mean, moving fast is impressive and important. I'm not trying to like, I, I, I, I think it's, it's really good. And maybe I'm wrong, and maybe those reactors will turn on and they're 90% capacity factor from day one, and that would be fantastic for the world. I'm not trying to say we shouldn't build high-temperature gas reactors. And I think it's a real good thing for the United States that we are moving much faster than what we used to do, which was take 20 years of planning before we could, and never even build the fucking thing. Yeah, like what, what Valor is doing something that is incredibly important for the space, even if I don't necessarily see the commercialization pathway in the same way that they do and their investors do, that, that culture of moving the industry faster is really, really important. But it— there's a certain level limit to what we can move faster on. And where I don't see us being able to trivially move faster on is getting the operational experience. And there's a mathematical reason why this is to get to that high capacity factor. So I'm like, this is not like, like, you know, Isaiah is an idiot. No, Isaiah is an extraordinarily impressive CEO and, and, and very admirable in what he's doing. And it's a— and like all startups, it's a bet. Mine's a bet too.

Yeah.

I just have a different perspective on what the market sees and how we're going to get to what the market needs. Yeah.

No, that's fair. How's your energy?

My, my, my—

how are you feeling?

I'm feeling great.

All right. Just got a few more. I mean, I've taken up a bunch of your time. No, no, no.

I, I, I, how's How's this going? I mean, like, I, this is, this is great, but I have, there's, yeah, I'm hoping it's not too discombobulated. Let me just make sure where we are. We're fine on time.

No, I love this. Okay. Is there, is there any SMR company, like, is there any SMR company that stands out to you as, as somehow more practical?

So I, I, I want to get back to this question of practical versus important.

I mean, is there anyone that's of them that stands out to you as having a better approach than the others.

So I mean, like, I think the GE, the BWX-300, like that reactor probably will operate at a high capacity factor. What I just don't— so there's, there's some basic questions that I always like to ask. It's like, why? You know, we've, we've built a lot of sodium reactors in the history of the world, and there's some commercially operating in the world. We've built a lot of small light water reactors, actually. I mean, the Chinese built built in SMR, right? What is different about this approach that others are taking versus what has historically led the industry to this path of large LWRs? And I don't really have a clear answer on that. So it's not as much like a critique of the, you know, it's, it's a weird, it's a weird, um, sort of question about like, what is the, from first principles, the better technology, right? Because that's a scientific question when we're really dealing with a real engineering challenge, right?

The—

there's very few things in the world that you think about that do what a power plant does, which is operate constantly 24/7 without stopping for 18 months or 24 months. Like, if you look at like a Merlin engine, on a Falcon, right? That thing has a mission duration of like 4 or 5 minutes and it does that incredibly well. And that's not a— that's not saying that what it does in 4 or 5 minutes is not an extremely challenging engineering problem that has been remarkably solved, but it is a difference in category from what a power plant system requires. It's why we have seen power plant and power engineering on the thermal system. Solar is different inherently from a physics perspective, um, than, uh, than what we see in other engineering realms. It's why so much— that's why General Electric, Pratt Whitney, and Rolls-Royce still dominate the global jet commercial jet engine space, right? It's why we see Siemens and GE, um, and Mitsubishi and Toshiba, you know, dominate the global power industry in, in these power cycle conversion spaces. It's actually very, very difficult to get these systems to work at the reliability that we need them to be competitive as electric power generators.

Okay. No, that's fair. And, and okay, you guys are located not far from Commonwealth Fusion in Massachusetts. I've been to Helion. Billions of dollars have been put into fusion as well. Fusion startups, you know, You know, there was this, obviously we've been here before, but kind of finally maybe somebody's gonna crack this nut and, and clearly some people believe something because all these things lit up all of a sudden over the last like 5, 6, 7, 8 years.

Sure.

Um, anything that excites you there?

So fusion is, is always exciting, but I think this is a place where, you know how I made that, that, that, that there's a difference between, you know, the campfire and the steam engine. We still haven't yet really figured out how to light the campfire. Fire and fusion. And just going back to first principles, you know, what is a nuclear fusion reaction? Well, essentially in the hydrogen-hydrogen fusion, we'll just keep it at that. We're taking two protons which are positively charged, therefore electrostatically repel each other. We're figuring out a way that they get close enough that they— that the strong nuclear force sticks them together and releases energy. That's a really, really hard thing to do that we haven't figured out how to do outside of, you know, a nuclear bomb, you know, a thermonuclear weapon or a tokamak or maybe the National Ignition Facility. You know, there are research facilities all around the world and it's only been just very recently that we even thermodynamically got more energy out of that process than we put it.

Right near here.

Yeah. At NIF. And of course that was not a, that was not a power generating design. That's, you know, something used to calibrate our nuclear weapons models. Right. And, and so we are still, at a point where it's very, very hard to figure out how we're going to sustain a nuclear fusion reaction that just makes more energy than we put in, which is like sort of square one of any power generating thing, is that you got to put more power out of it, you know, get more power out of it than you put into it. Uh, that's what, uh, that's what this is, right? Um, and so we're— we just haven't solved that problem yet. And I'm glad there's a lot of people working on it. But the prospects for that, you know, we're betting on breakthroughs, and betting on breakthroughs in science is something that obviously is necessary for progress to happen. But you can't bet on breakthroughs if we need power in the next 2 or 3 years. They're just too rare.

But like, what happened that all of a sudden there's multiple fusion startups in the span of a few years? Was there some sort of— it was just this moment in time when we needed energy and people got excited it again, or, or—

I, I, I'm not exa— I mean, I, I'm sure that there have been, you know, there's been a lot of innovations in superconductor technology. There's been a lot of innovations, right, in being able to model the plasma systems that are required to, which is a very complex magneto-hydrodynamic essentially problem that, that has happened. I, I'm not, you know, I'm not the expert to understand exactly why venture capitalists have decided that there's been a huge breakthrough here. But once again, you know, let's see when we first get a watt of electricity out of a fusion reaction, and then maybe I can give you a better answer than I can give you right now.

Okay. Another part of this discussion that I know nothing about is the uranium side of all this. Where is the US in terms of, say, say we go nuclear hog wild and we're building reactors all the time. I mean, isn't Russia sort of the dominant source for uranium?

No, for enrichment services, not for uranium itself. Russia has relatively little uranium relative to, you know, the biggest, you know, the 3 biggest countries in terms of proven uranium reserves is Australia, Kazakhstan, and Canada. But 2 of the 3, 1.5 allies. Yeah. Yeah. Well, I would argue that the Canadians are our allies, right?

Something— they don't seem happy with us right now, but that's why I have—

they still, you know, I bought maple syrup recently and it was from Canada. It didn't seem like a problem.

No.

So there's a lot of uranium in the world, right? Even the just the seawater. There's enough uranium to power the world for thousands of years. And if we really went hog wild on it, remember when we discharge a nuclear fuel assembly from a nuclear power plant today, 96% of the energy is left there and we have the technology, we know how to get that fuel out, to recycle it and reprocess it and reutilize it. I don't think it's particularly economical right now because uranium is cheap, relatively speaking, it's cheaper, at least in reprocessing. Maybe that will change. So I think we have and we certainly have enough uranium in the country also for a good amount of time. It's just not probably the most economic. So the Australian resources, the Kazakh resources, the Canadian resources are much more economic. Economical to retrieve right now.

But, but the enrichment part is—

but enrichment, right? So enrichment, we can't— if it's just a matter of stupid government choices that we didn't make that enrichment capability. You know, the French and the, and the Dutch and the Germans and the English have— you know, you can basically build, as there is right now, a centrifuge plant out in New Mexico that makes about one-third of the U.S. demand for enrichment services. We can expand that. That plant's expanding. The French are expanding their plant. The French are putting a plant in Tennessee that to serve a very good amount of it. Yurenka— sorry, Centrus, which has a totally American centrifuge technology, can expand in a building that's already built that could build 3.8 million separative work units, or roughly a fifth, more than a fifth of the current demand. And then you have new startups like General Matter, Scott Nolan's company, that can solve the enrichment problem. I mean, enrichment problem is solvable, right? This is mature technology that has been existing since the 1970s sort of gas centrifuges. It's just a matter of us choosing to prioritize that and building the national policy. And we've seen movement in both administrations really to get us off of foreign enrichment supplies.

And then like waste storage. I had never done a story on Yucca Mountain before until I did one on this company, Deep Impact, or Deep Isolation, that was trying to bury it in these fracking boreholes. And as I dug into Yucca Mountain. I mean, it was, it was, it was pretty like mind-boggling. I mean, my, my one takeaway from that was even if we have like good intentions around nuclear and we want to build again and stuff, it just seems so surrounded by just bureaucracy and fear and inactivity that even, even the best laid plans feel almost impossible. I mean, this is, this is like this storage center that, I mean, it's just as far as I could tell, will just like never actually be.

Yeah, I would, I would be, I would be wrong. I'm not sure I would bet on Yucca Mountain coming back. Yeah, but I don't think it matters. And I, and I say that literally not as a cop out, but as a serious policy perspective. You know, every nuclear power plant today stores their waste at the site and none of them are running out of storage space. Right. The amount, because nuclear fuel is so energy dense that literally the total amount of spent nuclear fuel that they ever generate can fit in a very small portion of their parking lot. And that's literally where most of it is. It's in these big dry casks that just sit out there. They're called ISFSIs. They just sort of are like the— and they just stand there. And, you know, this is a problem that procrastinating works on. The fuel only gets less radiotoxic with each passing year. It becomes easier to dispose with each passing year. And I do think eventually we will figure out a deep geological repository. You know, Finland and Sweden are building them right now, right? And they are, you know, they're EU, European, you know, uh, countries that are— I'm sure there's a lot of fretting and lattes and Volvo driving going on in that process, but they are getting there to the point. And I don't see that this is a, uh, a something that needs to be immediately solved, nor do I actually think— and I say this genuinely— I don't necessarily think it's something that we want to solve right now. Why do I say that? Because as I just mentioned, spent nuclear fuel contains right now in those dry casks outside those plants, 96% of the energy that it could possibly generate. In a world in which we have a breakthrough or an innovation in reprocessing technology, or even want to isolate some other useful isotopes out of that fuel, it makes a hell of a lot more sense to have that kind of hanging out in the parking lot at Palo Verde than it does for us to bury it thousands of feet underground, seal it up, grout it up, cover it up, and then have to get it all out again.

These—

this fuel, it may be not valuable today, but I could totally see in 20 to 30 years it being valuable. And my question to you is, why do we want to isolate it. We have never in the history of commercial nuclear power had a leak of high-level nuclear waste that has impacted anyone. Why is that? Because the fuel is solid metal, right? It's solid ceramic pellets encased in welded metal rods that are encased in big concrete bunkers. That fuel is not going anywhere. And in fact, we know from, from, you know, science that nuclear waste is not particularly mobile in the biosphere. You know, a billion years ago plus, there's a natural nuclear reactor. You know, billions of years ago, the concentration of uranium-235 was high enough that just nuclear uranium ore could fission if it was mixed with water and in the right geometry. And of course, when it fissioned, it created nuclear waste like any nuclear reactor does. Of course, you know, there was no cask, there's no storage. We we know from this, from, from that natural nuclear reactor, it's called Oklo, not the start— the startup is named after it, or not startup now, but, but sodium reactor company. We know that that fuel did not migrate more than 10 centimeters over a billion years without any protection. Right. It was literally just like uranium ore hanging out. And that high-level nuclear waste that it naturally generated never went anywhere for a billion years, which is thousands of times longer than we need for that fuel to decay to safe levels. So I once again don't really think this is a genuine issue. And I think that storing this in dry, above-ground dry-cast storage where it has almost no to literally no public health or safety threat is the right pathway because in 50 years or in 20 years, we might want that fuel fuel for something else.

Okay, 3 last questions.

Can you do it? I can do it. I, I, you know, give me another cup of coffee, we can go for a couple more hours.

There's, uh, I mean, 2 of the smartest, richest people I've ever met are Bill Gates and Nathan Myhrvold. For I think almost like 20 years, they've been really backing this TerraPower, and it's been their baby, and it's gone through like all these different iterations. What I remember when it first started, it was supposed to be taking nuclear waste.

Yeah, the traveling wave Yeah.

And reusing the stuff that everyone else was.

And now they're making, you know, a version of EPR too.

And, you know, they would tell me— and again, this is different political generations ago— but that, you know, they tried to do nuclear stuff in the US. It was impossible. They just took all their stuff to China. It was way easier to do it there.

They were trying to take the Westinghouse route.

Okay. Okay. I mean, what is the story on TerraPower?

I mean, I mean, I mean, I'm joking around, but TerraPower Power just got its construction permit approved by the Nuclear Regulatory Commission. Huge congratulations to that team. And they're building a, a, a high— what we call a liquid metal fast breeder reactor, in this case, liquid sodium. Right. They're not using it to breed fuel, but they're using a liquid metal. So rather than using water as a coolant, they use liquid sodium in a loop configuration or in a pool configuration. Actually, I was saying something else, but it's a pool reactor. This is similar to reactors like EBR-2 that operated in the United States and is similar to the Russian reactors at Beloyarsk, Beloyarsk-3 and Unit 3 and Unit 4. Obviously, there's different technical differences, but the general class of reactor is similar and they're going to— it looks like they're going to build it in Wyoming and we'll see.

And but it's, you know, when they first started, seems like a more novel pitch than what you're describing, which sounds like building something that we knew how to build before.

I mean, once again, you know, that's like saying like building a new— like building an AP1000 is the same as building a Westinghouse 3 or 4 loop. These are variations of the general reactor type of technology. There's definitely innovations in what TerraPower is doing. I think the— and this is more of an open-ended question— I have never seen any compelling case as to why those reactors will be more cost competitive than a large light water reactor. And we see this, right? The Russians have been building this type of reactor, the sodium fast breeder reactor, since the 1960s, continuously and commercially. Right. They built Shevchenko. Right. Which was in Kazakhstan, now in Kazakhstan. Right. Then they built Beloyarsk-3, a 600-megawatt reactor, then built Beloyarsk-4, an 800-megawatt reactor, and were planning to build Beloyarsk-5,, which was going to be a 1,200-megawatt reactor. And they made the conclusion that they could never even at 1,200 megawatts where they got a lot of economies of scale, they could still never get the economics to be competitive with a light water, with their light water reactor technology, which is a type of pressurized water reactor called the VVER, which they're building about a dozen around the world. And my question to TerraPower is a very simple one. What is the, you know, distinguishing factors of your design that makes the economics superior to the Russian reactor technology that had been attempted and is not able to be competitive with their large light water reactor fleet, given the decades of commercial development that has happened, the huge amount of R&D that the Russians have put into it, and just the experience they have in building them. That's my question. And it's It's not just, you know, not just them. The French tried to do this, right, with Superphénix and Phénix and Rhapsody. The Japanese did it with Monju, right? The, the Brits tried to do— I'm forgetting their reactor technology, their reactor name. It was not Dragon. That's their Dornier. You know, my basic question is, what is— what's the— what's the delta here? Like, what is— what are you doing that, that can really come over the inherent challenges of that reactor technology, which are are significant. You know, liquid sodium is a hard thing to kind of deal with.

All right. We'll leave that one for Nathan. Bill might be occupied for a while.

Maybe not. I won't.

What's the— to have some—

and talk about Nathan, about, you know, modernist cuisines, a magnum opus.

So, you know, I love talking to Nathan. Yeah. To have some tiny fraction of the knowledge you do, what is the the single best book somebody could read to ground themselves here.

You know, I find the— this is a terrible term— the lay literature on nuclear being really kind of poor. I mean, there's like, there's like good books like the, like, you know, if you want to read like about like Chernobyl, there's good Chernobyl books. But like, I find from an engineering perspective, like, you really do have to read the some of the more technical documentation, which one of the, you know, the NRC has a lot of bad things about it. But one of the good things about the Nuclear Regulatory Commission is that their entire document management system is public called ADAMS. And if you want to read about, you know, how the Fort St. Vrain, that high temperature gas reactor was designed in the 1970s, I would highly recommend reading the Fort St. Vrain Final Safety Analysis Report. Right. Or a preliminary safety analysis report just to get an idea of how these these plants are designed and some of the operational experience and incident reports that happen that were filed by the engineers. I personally am a very big fan of reading the design certification documentation of the various different reactors that are out there. I think for reading about TerraPower's reactor, reading the preliminary safety application report, the PSAR, is a really good place to start in understanding what actually does this design look like. Reading some of the operational reports from, you know, reading— if you're interested in molten salt reactors, reading Oak Ridge National Laboratory Working Paper 5018 is a really good place to start on that. The molten salt reactor experiment team about what they see as necessary to go from an experiment to a commercial power plant. There's a curated list here, but I guess what I'm trying to say is, is the devil in this technology The beauty is in the details. The beauty of nuclear power, the beauty of nuclear engineering is, is that it is something that is working on so many scales at once. You know, we are literally working at the subatomic level and we're going up to the macro-scale phenomenon of how in a light water reactor, a containment blows down with a massive amount of water and liquid. And so the— it is really technically challenging, but it's beautiful in that way. And the final thing I will, I will just say is there's a certain magic here that I don't want to get lost. And it's sort of why I have a political particular, um, perspective on this. But I want it to be clear that what all of us are doing, whether you're Valor, whether you're us, whether you're a researcher at Idaho, we are the pioneers of a totally new way, a literally new fundamental force of nature, of capturing that for man and using it to power civilization. And the depth of the challenge of that is just so immense and so beautiful, of understanding, you know, the quantum chromodynamics of how the strong force actually works, all the way up to the metallurgy and chemistry of making a new nuclear fuel, to the, uh, construction, packaging, and sequencing, and how do you get enough welders out there to build this. This is the deepest, most multifaceted sort of challenge because what we're doing is something that is inherently civilizational on scale.

It is literally—

we have— we were in the fire age. How are we going to go into the atomic age? And we're still very, very early in that chapter of human history, which is why I think going to the original sources, going to the original engineering, going to The original science is so imperative.

Well, that was fucking inspiring and way better answer than I ever would've given to that question. That was impressive. How many times have you watched Chernobyl?

The HBO show? Yeah. I don't think I've ever finished it. Not because like, not because I think it's like, like it's like a good, like I just like the, there's like these middle like episode, like I think like episode 2 or episode 3 where they're like shooting a cow and I'm just like, I'm so fucking bored.

Yeah.

Like, I, like, I eat a hamburger. I'm not like really that appalled by a, by a shooting. However, I would recommend the book Midnight in Chernobyl by Adam Higginbotham. Fantastic book. Really, really good on the, you know, the technical stuff. It's like it reads like a novel. It also goes into something that I'm very passionate about, which is like the Soviet nuclear industry, how that was organized, because it was like a total totally different approach. It ended up weirdly like convergent evolution on some of the same technical solutions that we have in the Western industry, but also how that entire industrial process was organized, how the towns came together. Really good book. I would, I would highly recommend it.

Okay, testing the depth of your nuclear trivia knowledge, um, which nuclear scientist brother co-founded the Exploratorium in, uh, J. Robert Oppenheimer.

Uh, yeah, it was J. Robert Oppenheimer's brother who was a, uh, was, was actually blacklisted as a communist and yada yada yada.

Frank, right?

Frank.

Frank. Correct. You passed. Amazing.

Yeah, I was hoping a little harder, but that's like a Wikipedia level.

I always thought, I always thought that was funny that he— it's like Robert Oppenheimer gets pitched as this, you know, in all these different ways. And it's very brooding and heavy.

There's other books. I'm sorry, you asked this question and I kind of gave—

Yeah, yeah, it's hard on this guy.

Read the preliminary safety analysis report of Fort St. Vrain. And like, yeah, it's a hard task to do. Not everyone's going to do that. Yeah, yeah. So like, if you want to, like, I would say the making of it, first of all, I would recommend Oppenheimer's biography. I think it's extraordinarily important book to read. Uh, Hyman Rickover's— any of Hyman Rickover's biographies are very, very important to understanding the role of the nuclear Navy and Rickover himself's, let's call it, acerbic and caustic personality in, in making the nuclear industry. I would recommend that. Um, The Making of the Atomic Bomb by Richard Rhodes is an extraordinarily important book to, to, you know, navigate yourself and, and, and, and read, um, uh, about— for, for understanding the, the Manhattan Project and how that led to the nuclear complex in the United States. Um, Red Atom, uh, which is about the Soviet nuclear industry, is a really good book. The Radiance of France, which is a book about the French nuclear industry, and how it sort of arose through industrial policy and how they actually achieved that, I would highly recommend. And there's like a couple more sort of more lay books that I would— read, Adam. I'm trying to think of others. I mean, just the biographies of Fermi, of Walter Zinn, these are all great books. Alvin Weinberg's books are also great. I, you know, I just— what I— what I'm cautioning about is these are great to understand the political and social forces that have driven the nuclear industry where it is today. But the engineering challenges of building a highly functioning nuclear power plant are different things, esoteric.

Yeah.

And they are really, really weird. They— like, I was just at a nuclear power plant literally last week, and we were like, the big discussion that we were talking about was the, um, the redox potential of feedwater, um, going into the final feedwater heater, right? And this has major implications for reliability stability of a component. And that seems crazy, like you're talking about literally the redox potential and like the, the, like the magnetite versus hematite. But what we're trying to do, if you think about what a nuclear power plant is doing, it's boiling just thousands and thousands of tons of water every day. And it turns out these really weird little factors of the metal choices that you're doing, right, about— they matter a lot. I mean, one of One of the things that brought— sorry if I'm going on and on, but just to give you an idea of how strange this industry can be, the entire French nuclear industry was ground to a halt in 2022 based on essentially how long a pipe was made in the industry in these new type of Frenchified reactors that led to thermal stratification, which put too much stress on a weld. Like the actual process designing these, these power plants is they are works of art because they require so many different aspects of chemistry, of physics, of instrumentation and control design, of welding, of material science that, that to— it is gonna be boring if you're not really passionate about it. You know, to write a book about the difference between Alloy 600 thermally treated versus Alloy 600 molynealed versus alloy 690 for steam generator tubes. That's not going to be a compelling page-turner, but it's the difference between billions and billions of dollars of economic activity that this industry can generate or not.

Okay. Okay. No, no, no. It's— I love that you put thought into these, these questions. Very last question. I think it's a decent one. The year is 2060. I don't know. You can change it if it helps. Answer. I mean, I just want to get a sense of where you think the US and China will be at in terms of percent of energy from nuclear, solar, more traditional forms. Yeah, can we do— can we—

I think it's going to be, uh, I, I think it's going to be— there's going to be a lot of— I think for a lot of places with good solar potential, I am, uh, you know, I think solar is a transformative energy technology for the world. And I think that solar is going to be, especially for places that are closer to the equator, it's gonna be absolutely one of the dominant, if not the dominant form of electrical generation and energy generation. I think though, for a lot of places where, you know, more northern or more southern, right, where just the, literally the rotation of the earth, the axis of the Earth prohibits— it makes it very challenging for solar to achieve high energy density, especially over the seasons. Nuclear will dominate. Now, I think there's, there's a really interesting question that we're kind of avoiding. If AI or some other technology generates a step change in the amount of energy that the world needs, like an order of magnitude increase, then nuclear will dominate. And perhaps that is what will happen. I mean, we're kind of flirting with that.

We just build a ton of reactors.

Because the problem is, is that solar inherently is limited by the insulation, the, the insulation, right? Like the amount of the, the solar flux, right? The amount of solar and photons literally per unit area per unit time. And you can't change that. There's, I mean, I mean, like, I guess you could make— move the Earth closer to the, you know, get around the inverse square law by moving the Earth closer to the— but that seems pretty impractical.

That's like Elon's next startup.

I— yeah, you're gonna need some nuclear power to do that, I would say, to say the least. I don't even think that's enough. But in a world where we're using 1,000x more energy per person, or 100x per person, nuclear is the only technology that exists that can actually generate that amount of energy, you know, short of paving the world over with solar or basically doing a lot of, I guess, orbital solar and then figuring out a way to transmit the power down, which is in my opinion, you know, death rays, you know, microwave death rays down, down to the Earth's surface. Right. So in a world where we really do scale energy consumption, the only technology that exists that can power that is nuclear fission, and if we figured out fusion.

So, okay, in that scenario, say 1000x more energy. Yeah, say this AI boom keeps going the way things are currently stacked up. I mean, China is very methodical and, and plans these things out and makes big countrywide decisions that, that last over time. Like, in that scenario you just played out, I mean, does the U.S. respond fund in the way it needs to, or is it just self-defeating?

No, I don't. I think we're capitalists ultimately, and I think that, um, maybe this is a little bit out of vogue right now, but I'm a capitalist. I do believe that ultimately capitalism is a superior system to state planning. And in a world in which that occurs, right, the market signal to drive for highly effective nuclear technology deployment will be so high, and already maybe is, we're starting to see it, hence why I'm in this space and raised a lot of money on a Series A, right? Um, we will beat the Chinese. I, I don't doubt that the United States has the capability to do so. Um, I do not doubt that the United States has the wherewithal to generate a massive amount of nuclear power. After all, this is an American technology through and through. Chicago, you know, December 2nd, 1942, that happened in Chicago, right? You know, the nuclear power was born. And it speaks volumes to me that the Chinese bread and butter nuclear power reactors that they're building dozens of, every single one of them has its roots back to American technology. We just We just need to refocus on it. And that's exactly what we're doing at Alva, right? One of the great things that I have the privilege of leading a company like Alva right now is taking the institutional knowledge that people that led that first nuclear buildup that built the world's largest to this day nuclear power plant fleet. I mean, what an amazing accomplishment. We stopped building in 1978. Still, the world hasn't caught up to us today. It still hasn't caught up to us. But there are still people around that have a massive amount of knowledge that live in this country from the reactor vendors, from the constructors. And I am building a team right now that is transmitting that knowledge down to the next generation and systematizing it, innovating on it, and using it as a foundation upon which we will dominate this century of nuclear energy. I have zero no doubt about that because our foundation is what the world has built on, whether it's French, whether it's China, whether it's Korea. They're all taking a US nuclear technology and improving it. And we have a possibility today for what I call boomerang tech transfer. We took the tech, we brought it from, from the US to France, and it's coming back from France to the US. Same thing with Korea, right? We are in a space right now where we have people who have done this, who know it, and there's an eager generation of younger people who are going to use new technologies to basically take their knowledge and bringing it up and make it 10 times better and 10 times faster and 10 times cheaper.

Well, I mean, again, that was inspirational. And, and, um, I mean, that seems like a good, good spot to stop. Yeah, we kind of came full circle, I think. And I, you know, I hope people, um, well, thank you for your time. I, I feel like anyone who listens to this will be much smarter on this topic than when they started. Um, and, and, you know, you're generous. You're, you're your mind. Look, I interview brilliant people for a living and, and I would say your mind's kind of unique, man. I mean, how deep you can, you can go on this stuff. So it was kind of you to spend this much time.

Thanks so much to us.

Um, thanks. And, um, we'll see you guys. Well, so what should people check out? Go to your website?

Yeah, go to our website. Um, but I, I would say more importantly than go to our website, sure. If you wanna, if you're an innovative driven person who wants to basically build real technology that's actually going to be deployed powering cities, and you want to learn from people who have done this before, who are, who have the knowledge, who've done tens of billions of dollars of engineering work in the nuclear industry, who have set world records in nuclear construction, and learn how to take what they're, they've done, learn directly from them, and work on applying modern technology to make it better, to improve on that foundation, come work for us. I mean, I think the biggest thing that this industry needs is it needs— I mean, we call it like sort of the sensei kohai sort of model. Like it needs young people who are going to challenge the status quo without abandoning the lessons that we learned from doing this and building the world's largest nuclear fleet. And that is what Alva is predicated on. It's built on that model of taking— we have, you know, people who have done this a lot in the company, and we have a lot of very young, innovative people saying, how do we actually use AI to, to expand that? How do we build a totally automated work packaging stack to monitor this? How do we make sure all of our virtual design and construction is done so that any change is integrated throughout all the work packages?

For example, reminds me a lot of SpaceX. I mean, that's kind of how SpaceX was born, was a bunch of 25-year-olds with a few people that have worked on the old Yeah. And it was like, you know, there's really not that much engineering you can do on a rocket. It's like we've known the physics for decades and it's like, can you make— can you modernize the electronics? Can you modernize the software? Can you guys think about—

I'm sure I'm not the first person who sat in this chair. So, you know, like SpaceX did, we're doing this exact thing. But the one thing I would say is, look, at the end of the day, the Merlin's an RP-LOX engine, right?

Yeah.

You know, and And that didn't mean, you know, they started out at RPLOX and then of course, uh, Raptor is methalox. And if you think about it, would they have been able to start the methalox engine without the experience and the engineering and manufacturing sort of knowledge that they got from the real world of building Merlin, improving Merlin, iterating it, firing it, launching it, using it? No.

No.

Yeah. Right. Scaling in hard tech is everything. Scaling on software is trivial. That's what's brilliant about software. But when we're talking about atoms, not bits, right? Actually figuring out how to scale stuff is sort of the secret. And I think that's what Alva's working on. It's just like literally the tech that we're building right now, and we're going to be constructing it and designing it and engineering it. That is our RP live. Blocks engine and we're bringing in the old Rocketdyne guys, in this case the old AECom guys, right, the old Bechtel guys, to, to basically teach us how to do it and improve on that. But that does lay the foundation for not only just incremental innovations but truly revolutionary evolution, uh, evolution in the long term and medium term.

That's a difficult balance, you know, because I feel like Blue Origin in the early days took too many of the old line.

Yeah, look at how slow they are. Yeah, yeah. I mean, let me tell you, it is my number challenge and our CTO's number one challenge is culture, right? Is making that balance work. Right. I think the difference is we lucked out in that the, the, that, you know, Rich Coleman, for example, our head of, of projects, right, who has this, you know, who's in the 60s, has $20-some-odd billion of construction, ran the largest N triple S modification EPC in in the world, right? There was a, a great moment when I was like, this is the exact cultural fit. And we were doing an engineering project and we need to do a concrete and steel takeoff. And Rich was like, okay, who do I go to, to, you know, do the takeoff? And we're like, well, you're kind of looking at the team here. Like, and he is like, okay. And then he literally by hand did the takeoff himself. And he said, and he probably hasn't done a takeoff since 1984, 1985. And there is a certain generation, and I think in, in building any of these companies where you're trying to basically take people who are, you know, much more experienced and who are not only willing to want to do things differently, they are eager to want to do things differently. They are looking at the fact that they don't have a, you know, a whole material takeoff team and department and procedure, and they're saying, finally, I can build it better the way that it should have been built in the first place.

That's the key. You need the experience. Experienced people who are deeply dissatisfied with, with the status quo, where they were. Yeah. Who, because they have to be willing to do that work, do it cheaper, do it faster. But, and they have to also be willing to kind of listen to the young people challenge them on some stuff, you know? And, and so, but that, it is, it's hard.

And, and maintaining that culture, right? We, we like to call it like creative tension, right? It has to be Well, um, um, the, the, there's two, we have two rules on this, right? The first is if people have been, it's never an argument to do something the same way, this way, because people, this is the way it's always been done. Rule number one. But rule number two is if things have been done this way over and over again, there is generally, but not always a very good reason why it's been done. And if you keep those two ideas in tension, what you'll find, what will emerge out of that, is sort of this Pareto Optima of what is building off of a lot— hundreds of billions of dollars of real-world experience— versus sort of separating things out that are just there because of cultural inertia and can be done better.

Yeah.

And I think that is what, you know, the cultural challenge that Alva has, and that any company that is taking an industry where the US was dominant in 30 years ago has sort of gone to sleep a little bit, and now we're trying to revive it, is recognizing that these people, these legends in the space, are your best resource. But you're going to need to be able to take those legends and build and innovate off of them, with them.

Yeah, yeah. No, no, that makes sense. Well, we will be watching, James. Thank you.

Thank you so much.

No, man, thank you for your time.

All right, bye-bye.

The Core Memory Podcast is hosted by me, Ashlee Vance, and/or Kylee Robison, or both of us together. It is produced by me and David Mickelson. Our theme song is by James Mercer and John Sortland, and the show is edited always by the John Sortland. Thank you so much to Brex and SendCutSend for all your support. And thank you most of all to everybody for listening or watching. We love you. Please leave us a like, a review, a subscribe, all those tremendous things. Thank you, and we'll see you again.