Ep 157: Scott Nolan on the AI Energy Crisis & America's Nuclear Renaissance

18:13Scott Nolan:We don't need to do this anymore. This is like the ultimate though, de-industrialization. We talk a lot about things we turned off in the US. We turned off all of our nuclear capability here in terms of in terms of enriching all the time that we kept somewhere and then for like the army or something or no uh so what the military has is left over from that era it's a big stockpile um and we shut down the last enrichment site in 2013 and that site was actually in western kentucky on doe land that our company is now building on so you're now so you're not going to build where that last site was in western kentucky that's right and and once you so you're giving this given this material and and you and you make it a gas and enrich it or maybe you're giving the gas and you enrich it and then you give it to somebody else once it's enriched because there's already a process for making the fuel pellets that's right the way this industry works the utilities take care of their own fuel they'll go buy the uranium out of the ground from a miner and then they will own title on that and they will buy services all the way through so they'll just for every for every little piece or you do this part you do this part you do this part so there's already people who are really good at making i mean shouldn't you make more of the new little tiny poppy seeds or something because there's going to be more plants and no one's doing it or it's not it's not worth doing that part so what else is doing there's there's a bunch of companies doing that.

19:20Scott Nolan:So I think that'll be taken care of. The step that I was most concerned would not be taken care of in time was enrichment. And so we're doing that step. And so to your question, the business looks like a big building. Think of like a warehouse, data center, biopharmaceutical plant, big rectangular building, lots of equipment inside. On one end comes in a big cylinder. Think of like the big propane cylinder behind people's homes in rural areas. on the other end, out goes two cylinders, one enriched, one not enriched. Amazing. And let's step back for a second. I was with Elon recently, he was here and one of his big concerns for the US was not having enough power.

19:58He's very impressed by how much power China's making overall. I think they're bringing on maybe like a new Texas worth of electrons every year, something like that. I think overall, even though we might be higher per capita, they're like 3X, I think America right now is this massive amount. And obviously for the AI race and for AI in general, and for, I think for the wellbeing of our average citizen in general, we want more and cheaper energy. And so the question is like, what's the answer? And I know he's focused on solar is one of his things. I know natural gas is something I'm pretty excited about.

20:26My friends are building more natural gas plants. We have tons of cheap gas in America. Like, like where does nuclear fit into this? Like what, how much of the solution is it? Is it a key part of the solution? How do you see it?

20:35Scott Nolan:Yeah. So I was talking with someone about this recently and The US has done an amazing job on actual fossil fuel production, fracking, natural gas, exports. And that's actually been the mechanism that we've used to bring down carbon emissions in the country during the last decade. Yeah, fracking and natural gas is way better for the environment than coal and oil and other things. Yeah, half the carbon per unit of energy, basically. Yep. And so you just emit much less carbon doing it that way. You know why fracking was so unpopular? Because Russia spent a bunch of money convincing Europe not to do it.

21:03This is one of those things that drives me crazy. I don't know if you know this. They funded the Green Party. They funded the protests. and they turned off fracking in Europe because it's so good for the environment. They tricked everyone because obviously they don't want competition. So I think it's probably also a nuclear thing too where people don't. It's nuclear competition as well. But anyway, go on.

21:18Scott Nolan:Yeah. Well, yeah. US has done a great job there. But to your point on the grid, actual electrons, like that's fuel. We can sell that, export it, use it for other things. But when we're talking about literally making electricity, which up until now is, you know, it's always been on the grid powering data centers. If you look at the grid, in 2010, we were completely tied neck and neck with China on grid total capacity. Something like 4 ,000 or 5 ,000 terawatt hours per year. Since then, in the 15 years since 2010, they've tripled. I think this year they're actually at 3x the U.S. grid capacity. And we just haven't got it.

21:54Basically flat. Why not? What are we doing? Is it just like, why couldn't we go up? Is it not worth investing? It probably is worth it financially, right? Or do we just not have that demand?

22:02Scott Nolan:It's worth it now. I think a lot of people thought for a long time that it was just about efficiency. It was about US moving out of these energy intensive industries into services industries. Oh, because we were de-industrializing. So China was building manufacturing. We were de-industrializing. We didn't need more power. Now we're realizing, wait a second, we're going to re-industrialize. So we need the power. Yeah. The conventional wisdom was GDP is completely correlated with energy consumption up to a point. And once you get to that point and you go past industrial sort of operations and you're doing services, businesses, then maybe you don't need as much.

22:34Scott Nolan:I think now that's completely flipped with data centers consuming so much power and needing them to scale the next few years. And EVs, advanced manufacturing, I think turns out a lot of manufacturing is going to come back on shore, but we need the power. We can't do it. Yeah. I think we need to turn around from viewing it as being okay that we've had a flat grid to realizing we need to almost go vertical on it. I mean, it's a disaster for our whole re-industrialization. This is a big theme for all of our stuff. You're a founder's fund and doing a bunch of things in these different areas, right?

23:06Like we need to re-industrialize. We need the power.

23:07Scott Nolan:Yeah, it's not even the amount of power. It's the cost of power. That's the other part. So that's why I think, you know, with nuclear, it's not just can we build more nuclear, but can we make it so cheap that we bring a lot of these things back to the US, like aluminum production, steel production, these things where the major cost is just electricity cost. Nuclear power could be a very big part of the production, you think? Like, what does that look like? Like, what's the cost of power right now? What could you do with mass nuclear production? Yeah, I mean, cost of power is like wildly different based on state.

23:36Scott Nolan:I think, yeah, what could it be? I think back in the, what was it, 60s or 70s, I think in today's dollars, nuclear was costing about three cents per kilowatt hour. That was the actual production cost. Obviously, in places like California, you know, people are paying 30 cents. So I think there's a lot of room there. Yeah. No, I think nuclear should be a huge part of the mix. Today, it's 20 % of the US grid, which a lot of people don't realize. So it's already still 20 % now, and you think it could be a lot more? Yeah, it's I think 18.6, 18.7 % right now. We're going to see some up rates the next couple of years, people taking plants that are already online and just amping them up with different fuel.

24:18Scott Nolan:And then I think we're going to see some restarts as well. So we'll see that grow a little bit. But you talk to any of the hyperscalers who want to build big data centers in the 2030s. And I think for all of them, it's about nuclear. How did you get this idea? You're also an investor at Founders Fund. You guys have been crushing a lot of areas. Were you looking to potentially invest in nuclear in different areas? This was a gap? Yeah. So over a decade of just looking at engineering driven companies, really starting in things like satellites, ended up going all the way across pretty much every space ending up in energy in the late 2010s, invest in a company called Crusoe, invest in a company called Panthalassa.

24:59Scott Nolan:These were really about stranded energy sources. And then invest in a company called Radiant that was looking at stranded demand, not stranded supply. What does that mean? What's stranded demand? So imagine an Alaskan village that goes entirely off of diesel and they get a shipment in the summer only when, when it's not iced over. Yeah. And so places where people have to pay a lot of, uh, you know, a high cost of electricity and where a small nuclear reactor can actually compete with the cost that they're already paying. So what companies are you bullish on providing those reactors, by the way, are there certain ones that you're more than others?

25:33Scott Nolan:I mean, I would say there's three segments. You have the micros, which are for applications like that. You've got SMRs, which can be anything from, uh, data centers to industrial processes. And then you've got the really big reactors gigawatt scale for grid. And so I think we need all three. I think all three are going to scale. When you asked before, like, oh, should we just do the SMRs? I think it's all of the above. We should do way more gigawatt scale, get really good at building them, build them the same way every single time, bring that cost down. And so then you've got SMRs that could be great for data centers and then the small ones.

26:06Scott Nolan:I think there's going to be winners in every single category. But to your point, how did I find out about this? it was meeting companies in the space for years. And the thing they all said was actually getting the license is not as hard as you think, Founders Fund. The hard part is getting the fuel. And they said, we're really hopeful that we can import some of this fuel, this HALU fuel. HALU stands for high assay, low enriched uranium. So that's the 10 to 20 % enrichment level, which everyone converges on 19.75 because it keeps you out of weapons grade, but helps you get the most density of energy in your reactor.

26:45Scott Nolan:And they said, we can't get the fuel. We have to import it from Russia. So the obvious question was, why don't the U.S. companies that make all the other fuel just make your specialty fuel? And the answer was basically like a blank stare back. Like, what do you mean other providers in the U.S.? There are no other providers. And so spent almost a year looking into that, understanding what's broken on the supply chain, what part is missing. And the answer was, it was the enrichment piece. We didn't have domestic US company led enrichment. And so tell us about, you raised a bunch of money for this company.

27:21Like what's, where is it, Ryan? What's this timeline? Yeah.

27:23Scott Nolan:So timeline is driven by the market. So there's two market dynamics. One is all these new advanced reactors need to not only prove, you know, that they work, prove their safety, prove their performance, prove cost structure. They need to scale. And so their initial bits of fuel are coming from DOE, but to scale deployments, they need new sources. There's already fuel the Department of Energy has to give them for their tests. Yeah. But as they start to scale, they're going to need you. So you're ready, you're saying? I think that timing is coming, you know, 2028 through 2030. They're going to first really start doing the, you know, single digit deployments.

28:00Scott Nolan:And then I think 2030 is where it really ramps. In 2030, you could potentially see a lot of these plants coming online, in which case the demand goes way up for you. And so our commitment is to be online by end of decade. Got it. So you need to rush right now. So we were talking earlier about what's the thing that's going to hold you back? What's the thing that's going to take the longest? Tell us about that. What are the things you're sprinting to get done? Yeah. So if you look historically at timelines, you would look to licensing as probably the timeline driver. What's that? Licensing what?

28:26Scott Nolan:So getting a license to operate the facility. And that comes from the NRC, the Nuclear Regulatory Commission. That used to be impossible to deal with. I'd imagine it's a little bit easier to deal with now or maybe not necessarily. Oh, it's become much easier. So, uh, you have to say that cause they have to give you a license. No, absolutely not. Uh, if you go back to, if you go back to 2018, um, and you ask those companies like, you know, is this the hard part? And they would say, no, it's not the hard part. It takes a little while and you have to do the work, but it's predictable. And what we're all trying to figure out is, you know, make sure that this is going to be a safe reactor.

28:58Scott Nolan:That's what everyone would have told you. Um, you know, I think today is actually the one year anniversary of the, of the nuclear executive orders. Um, and so there were four executive orders covering DOE, NRC, DOD, and the supply chain that has really created a huge tailwind for the space. So if we go back to like 2020, people would have told you it's a very, uh, you know, clear process of what we have to do. The timeline is unclear. If you then go to a week or two ago, you might say, okay, the executive orders, you know, did say, did give some timeline limits on reactors. They did not give clear timeline limits on fuel cycle facilities like ours.

29:41Scott Nolan:Yep. But you could say a fuel cycle facility is maybe equally difficult or some, you know, less difficult than a reactor. And so if they're going to do 18 months, turn around on reactor license applications, you would maybe guess something like that. And if he gets 18 months, then it can end up being the long schedule driver. Yeah. And I think it was Thursday, the NRC announced that a new fuel cycle license application went in and they're committing to put it on fast track and get it out in 12 months. There's someone other than you who's doing that application. Yeah. And so if you assume 12 months, all of a sudden it's back on the companies.

30:21Scott Nolan:And now the companies are actually the timeline driver. And so it becomes one of these like types of very difficult company called that, you know, Peter at Founders Fund would call a complex coordination company where you're working on eight different work streams at the same time to go as fast as possible. If any single one fails, it will delay the project. And so the key capability is how do you just coordinate everything? And so teach us a little bit more. One of the work streams was that license which you haven't, you've not applied for yet because you know, because you have to do certain things first.

30:49Like what are your main work streams?

30:51Scott Nolan:Yeah. So one work stream is license. So, So submitting that this year, another work stream would be technology, just improving the technology's performance relative to cost. Another one is construction. This is one people really underweight, but here in Austin, you guys have seen it firsthand. We like building fast here. Our friend Omid, right? They did that fast with Elon, the plane. Yep. Austin Gigafactory. And so the key thing on construction we've found is you can't just hand it off to some other company. No, no, no. You have to own this. Ceronics doing all of its own stuff. We're doing thousands of autonomous boats here.

31:24You got to do it yourself. You got to do it yourself.

31:26Scott Nolan:Yep. So this is classically - It's part of the core IP. It's classically called EPC, Engineering Procurement Construction. And so just like those companies, we build an in-house EPC. It's about 20 people in Kentucky. They're leading our project. They'll work with the subcontractors, the welders, the grading firms, you know, cement, everything. But it's run in-house. You have your project managers in-house. It goes so much faster if you do this right. Yeah. So construction and licensing and the technology itself, those are the main things. Manufacturing. What are you manufacturing? The hardware that would go inside the facility.

32:00So you got to design it with the science, then you actually have to manufacture it all.

32:03Scott Nolan:Exactly. So it becomes a scale manufacturing problem. So it's like, you know, Starlink, we've got a user terminal factory here in Bastrop, what, 30 minutes away. And that's a problem where, yeah, you can build one, but can you build a million of these things? Can you build a million a year? How do you scale that? That's the hard part. And then how much money do you need? I think you also got like a massive amount of money in Exxon financing, right? Because this is key for the US. So not only did Trump speed up the licensing, he's also like putting money into this area for cheap loans or what? How does that work?

32:31Scott Nolan:Yeah. So the funding we've gotten spans venture capital equity, debt on things like manufacturing equipment. There's a grant we received from the DOE. It's really an IDIQ, so indefinite duration, indefinite quantity contract, in which it's a lot like NASA's COTS program, if you remember back to that with SpaceX, where it was completely milestone based. So you do certain things, you unlock capital. So you prove you're doing it right, they'll give you more money. And not cost plus, but fixed price. And so our whole contract with the DOE, which we received in January, we received that award under President Trump's DOE.

33:12Scott Nolan:actually were down selected and first applied under the Biden administration. It was Congress that appropriated$2.7 billion for HALU enrichment. And we received one of those awards for 900 million. So$900 million milestone based contract, which will be earned over the next few years as we build that capability in Kentucky. So that's part one. Part two is what you referred to on the EXIM side. So the Export-Import Bank of the U.S. wants the U.S. to be competitive internationally, be able to play on the same level playing field as other countries and create jobs in the U.S. And so we have an LOI with them to wrap our contracts that we sell into Japan and Korea to help pull some of that financing forward to help us build a greater scale.

34:01Scott Nolan:And so right now that's an LOI to go do those contracts. And then we're working with Japan and Korean utilities to move that ball down the field. That's cool. Japan also has like hundreds of billions to invest in U.S. energy. I guess you should probably talk to those guys too. Yeah, yeah, that's right. I think the XM deal is really just around utility contracts. And so that's the next big piece. We're not just interested in that as something to expand the commercial scope of the business to move internationally. there's a big U.S. national security benefit of that, or like even a global non-proliferation security benefit of that, which is if the U.S.

34:45Scott Nolan:does not provide enrichment to these countries, they will have to get it somewhere else as they build up their nuclear fleet, or they turn back on their nuclear fleet. And that'll be Russia, China. So yeah, you want to be part of the U.S. kind of power group, not the other guys. Right. That seems really critical for the U.S. to get this right. and have you thought about like what's the trade-off a lot of people would think oh if you could do this you might as well be the one making like tons of the power yourself and owning a big utility but a lot of people say no just stay really focused on the thing you're doing it feels like more like what elon and some of our other successful entrepreneurs we know would do to own the whole thing yourself versus just do the one piece but there's something about just the one piece you got to ace i guess for now yeah i mean this is always um thinking back to like the very beginnings of my time at founders fund it was working on peter's class cs183 yeah that became the book zero to one and one of the core lessons from that was you know don't think of this as don't think of the energy market as a trillion dollar market and oh if only i can get one percent of that trillion dollar market it's much more focus on something that you can do on the narrow piece first yeah focus on the really narrow thing where that no one's meeting the need and where you can actually deliver huge value to the industry and where you can you can just perfect that and then you expand out once you win that who knows you could start building other things being more aggressive.

36:01That's right.

36:01Scott Nolan:And I think, fortunately, in many other parts of the supply chain, there's dozens of reactor companies. There's a bunch of people making the fuel pellets. I think the thing that we want to solve is the enrichment piece. And if we do that, we think that alone can be a huge business. It definitely can be. So stepping back a little bit, let's look ahead. What are the new possibilities for nuclear otherwise? I think Jared Isaacman, who is on this show, runs NASA. He wants to maybe do nuclear power in space. Is that something you think this stuff's going to happen? Absolutely. Yeah. I was just reading some of the announcements.

36:29Scott Nolan:I think they're aiming for nuclear-driven space propulsion by 2028. And then they're talking about reactors on the moon's surface to create heat and power things. It's great. I think it's, you think about like, if you just think, how are we going to power things on the moon, especially the dark side of the moon? Nuclear just makes the most sense by far. You just got to make sure it doesn't blow up on the way up there, cause a lot of trouble. Yeah, true. We have some pretty reliable rockets these days, fortunately. And I guess moon aside, what does it mean to have like energy abundance on our planet?

37:06If we can make energy as cheap, go back to three cents everywhere or whatever, or there's much cheaper everywhere, you know, what does that mean for us?

37:13Scott Nolan:Yeah, I think three cents shouldn't even be the limit. Like that was three cents building huge gigawatt scale mega construction projects, not factory production. And so I think we should physically be able to get less than three cents if we were doing it before. and you start talking about energy that cheap, there was the classic phrase, too cheap to meter, which I don't think, maybe it doesn't have to be so cheap that you can't even afford to put a meter on the house. I think that might not really make that much sense. But so cheap that you can bring back all these industrial processes to the US, you can rebuild the manufacturing base and just have immense economic growth.

37:51Scott Nolan:I think that's exactly what happens. I think it's kind of fun if energy is really, really cheap and AI is really, really good and it starts working in the physical world, then basically anyone in the middle class is going to be able to build this, like, you know, a giant cave that they designed. It's like 40 moons underground, right? Or whatever. I just think about it. It's like crazy, like, or a beautiful castle. You're going to be able to have castles being designed and built by everyone if they want. I mean, probably HOA should have ruled. Yeah. As long as HOA is okay with this underground.

38:18I don't necessarily want giant castles everywhere I live, but you can go buy your own land. You do it. It's just going to be amazing. Everyone's going to be able to build things no one could have built before, right?

38:25Scott Nolan:That's the idea. Or just that we get so much more prosperity. Yeah, everything should be cheap, right? Everything should be a lot cheaper for everyone. Yep. I love it. Well, I'm very bullish on the future. We started this podcast actually to push back on cynics and doomers. Like what makes you optimistic for America? Somebody was reminding me that this year is America's 250th anniversary. And they said, okay, what do you think about the next 250 years? And I thought back and, you know, you look up, okay, 1776, a long time ago, what do we have? We didn't have electricity. We didn't have cars. We didn't have antibiotics.

38:56Scott Nolan:We didn't have any of the stuff we have today. And so you fast forward like another 250 years and you say things are actually accelerating. What's it going to look like? I think we're going to have the science fiction future that everybody was promised as long as we don't stop it. I think today we'll look even more backwards 250 years from now. Than the last 250 years. I think people don't realize that. Like they'll see us and they'll he has these like crazy barbaric ancestors that are like just everything's like completely weird and different and hard yeah that's my view i think it's gonna be exciting i think the next 10 years we're gonna see a lot of this stuff really take off just gonna make sure everyone stays tough yeah because you know if everything's too easy it's like it's fair fair the cycle but yeah in general i'm very optimistic though well thanks scott for joining us yeah thank you