1.
At 27 years old, with a baby face and a receding hairline, Isaiah Taylor looks like nothing so much as a very large cherub. After dropping out of high school, he launched into entrepreneurship; he has described himself in his professional bio as a “self-taught engineer and 3x founder.” The first two companies were an auto repair shop in northern Idaho and a software system to allow auto repair shops to track the condition of their customers’ vehicles. The third was a nuclear energy startup, Valar Atomics, with hundreds of millions in capital, a factory in El Segundo, California, and a very active social media presence. (Taylor tweets regularly: pictures of him smiling next to the red Tesla that Trump bought from Elon Musk before their falling-out; paeans to God, “the empire,” and “Western civilization”; and more scattered thoughts, like gratitude for a national nuclear laboratory: “Fizz fizz. Fizz fizz. Uranium so good! Thank you Oak Ridge!”)
Taylor founded Valar in 2023. He has said he pitched his company to some 80 different venture capital firms before Stephen Marcus of Riot Ventures gave him his first investment. That was, frankly, a crazy bet: Taylor was only 24 years old and had no real connection to the nuclear industry, apart from a paper brief on his vision. Last year, the bet paid off. In February, Valar announced it had raised $19 million in seed funding and unveiled its first reactor prototype. Then, on May 23, Donald Trump issued four executive orders that have transformed the U.S. nuclear industry. These called for new public subsidies across the entire sector—from enrichment to plant construction to the disposal of radioactive waste. Crucially for startups like Valar, the executive orders also outlined regulatory transformations that would allow companies to build small reactors, load them with fuel, and turn them on without having to go through the painstaking licensing process of the Nuclear Regulatory Commission.
As news of Trump’s orders broke, Taylor published a manifesto heaping praise on them. (“There’s a new arm to national nuclear security: Dominance. Dominance in civilian nuclear technology development, dominance in nuclear energy infrastructure deployment, dominance in shaping global development.”) The same day, Taylor went live on Bloomberg TV. Alongside Utah Governor Spencer Cox, the young CEO announced that Valar had signed a deal with the state to build an advanced reactor there that would be operational by July 4, 2026. “That’s what the president has asked for,” said Cox. “It’s absolutely possible that we can do that.”
The timeline is immensely ambitious. In a 2021 study (from Oak Ridge National Laboratory, actually), researchers looked at how long it took to build over 500 advanced research reactors “from first concrete pour to criticality” with appropriate safeguards. They found that a majority had taken at least a year to build, with the average time being 32 months. Valar, as well as a handful of other companies selected for the Department of Energy’s Reactor Pilot Program, are attempting to do the same thing in a fraction of the time. The DOE maintains that three companies are on track to turn something on by the president’s deadline, although it is cagey about which companies exactly. Valar is gunning to be one of them.
Some critics have questioned the wisdom and purpose of this breakneck sprint. Paul Dickman, a retired senior policy fellow at Argonne National Laboratory and an adviser to the Japanese government on the decommissioning of the Fukushima Daiichi reactor complex, called it “bullshit” when I spoke with him. “I always tell people I don’t need to wait until July Fourth. I can do it tomorrow. I’m gonna go down to PetSmart and get myself a fish tank. I get myself a California source and a piece of fuel and I’ll have criticality tomorrow,” he said. “Of course I have a lot of dead fish floating around my fish tank, but that’s OK, you know.”
Others have pointed out that the United States has no long-term solution for waste disposal. Or that major questions hang over the economic viability of the small modular reactors most of these companies are building. Or that the reforms Trump has enacted at the Nuclear Regulatory Commission look like regulatory capture. Even further afield, there are those who view the current bipartisan enthusiasm for nuclear energy as a pernicious distraction, given that almost none of these reactors will come online soon enough to service the data-center boom or affect global carbon output in time to evade catastrophic climate change. “The first thing to understand is there isn’t much of a there there,” Allison MacFarlane, director of the University of British Columbia’s School of Public Policy and Global Affairs and former chair of the NRC, told me. “None of these things exist, OK. You can’t go and buy one and have it built tomorrow or even probably 10 years from now. So that’s the reality.”
Thus far these voices have been little more than a distant chorus to the forward march of industry. Asked recently what success looks like for the NRC, Ho Nieh, whom Trump appointed as NRC chair in January, replied, “Shovels in the ground.”
2.
I first spoke with Taylor in summer 2025, a few weeks after Trump’s executive orders were announced. He popped up on my computer screen seated in a rattan chair and ready to give me his pitch. “Most of the time when we’re talking about building reactors, these are like five- to 10-year research projects, which maybe happen, maybe don’t,” he said. “And my whole philosophy in starting the company was like, we have to start moving faster as a country.” China, which had started building out a major domestic nuclear industry only this century, was on pace to overtake the United States in nuclear energy generation within a matter of years. It would require “a massive leap” to catch up. He thought Valar could do it.
Part of the reason I had been interested in Taylor and Valar was that they were such outliers in the field. Taylor has a great-grandfather who worked on the Manhattan Project, but his childhood was spent following his own dad from state to state as he chased white-collar sales work and the like. He says he grew up on food stamps. Their car was once stolen by a family friend, whom they confronted and forgave. I found these details immensely sympathetic when I heard Taylor relate them in an unusually personal interview he gave to the podcaster Shawn Ryan. I felt the same way hearing Taylor speak about his mother’s intelligence and how she used to discuss physics with him when he was a child.
All this cut against some other salient facts of Taylor’s life, which reporters in Salt Lake had been writing about of late, after his company announced it would build a nuclear reactor in their state. Like our secretary of war, Pete Hegseth, Taylor is a member of Christ Church, an institution that was founded and is still run by a pastor named Doug Wilson. Wilson wants an America in which non-Christians would be barred from public office. In a tweet about Wilson, Taylor said he appreciates the pastor’s teaching on “Christian wealth.” For Taylor, that not only means money, but also friends and family and other forms of wealth, although money is a big piece of it. (“Certain exceptions aside, participating in the system of wealth creation is simply blessing your neighbor at scale.”)
More directly related to what Valar was attempting, Taylor had erroneously claimed in a press release posted to X that you could hold spent fuel from his reactor after it had been removed. (“Nuclear engineer here. This statement cannot possibly be true,” Nick Touran, a prominent nuclear commentator and indeed a nuclear engineer, replied to the tweet. Fuel from the kind of reactor Taylor was talking about “would give a person a fatal dose within a few seconds if they were to hold a handful.”) And there was the unfortunate fact that in 2023, months after Taylor founded Valar, his friend and director of business operations, Elijah Froh, had sued Taylor’s other friend and head of operations, Kip Mock, for pouring diesel in a wood-burning stove and inadvertently setting Elijah on fire.
When Taylor and I talked, we focused on his criticisms of the Nuclear Regulatory Commission. Like most leading nuclear startups today, Valar is pursuing a small modular reactor, or SMR. Its chosen design is cooled with helium gas, and Taylor has called it “the Toyota Camry” of nuclear reactors. (That has to be understood as a proleptic description, as there is currently only one commercial version of such a reactor in operation in the world, and it is in Shandong, China). Also like most of its competitors, Valar has a business model that leans heavily on the notion that it will build its reactors in a factory. For years now, analysts have suggested that bringing construction inside a factory could help avoid the cost and schedule overruns for which the nuclear industry has become notorious. There is the tantalizing likelihood, too, that repeated construction will yield major efficiency gains, as mass production has tended to do for most products. Taylor is particularly captivated by these prospects. He envisions hundreds of reactors rolling off Valar’s assembly line every year, populating huge groupings of reactors that Valar calls “gigasites,” and possibly, at some point in the future, being installed on Martian soil. The primary obstacle standing in the way of such a future, he explained to me, was the “regulatory matrix.”
In April 2025, Valar had joined two other nuclear startups and the states of Texas, Utah, Louisiana, Arizona, and Florida as a plaintiff in a complaint against the NRC. Their case hinged on the claim that the small modular reactors that Valar and other companies planned to build posed “no meaningful risk to ‘the health and safety of the public.’” Because of that, the plaintiff’s lawyer argued, these reactors did not fall under NRC oversight. There was some exegesis of the Atomic Energy Act of 1954 involved, but in the main, the suit was asking a judge to adjudicate the basic safety of a broad category of nuclear reactors. To me, the whole thing seemed insane on its face. A report from New York University’s Institute for Policy Integrity also points out the risk of a “fifty-state patchwork of separate licensing regimes” if regulatory authority were taken from the federal government. But working on the rough heuristic that the Supreme Court had systematically undercut the authority of federal regulators over the past half decade, and that the suit against the NRC was being heard by a member of the Federalist Society, I reckoned Valar and its co-plaintiffs had a reasonable chance of success.
Early in our call, Taylor wanted to show me a chart. “So this is the cumulative U.S. nuclear construction permits over time with Three Mile Island drawn in,” he said. What that looked like on the page was a yellow line ramping upward at a healthy rate from 1955 until 1979, where it was bisected by a vertical red line marking what for Taylor was a diluvian event. That year, in March, a broken valve in the Unit 2 reactor at the Three Mile Island nuclear power plant precipitated a partial meltdown of the core and the release of a plume of radioactive fission products into the surrounding area. No deaths were directly linked to the disaster, but the U.S. nuclear industry never recovered. On Taylor’s chart, the yellow line effectively flatlined after this point.
There are a host of competing interpretations of exactly what went wrong with the nuclear industry over the 1970s. Rising construction costs, stagnant energy demand, and an ascendant anti-nuclear movement all cut into nuclear’s bottom line over the decade. By 1978, the industry recognized a crisis. Then came Three Mile Island. No new construction permits were issued for the rest of the century. In the past 20 years, the NRC has authorized the construction of 18 reactors (a fact that Taylor’s chart gestured at, if you looked closely), but only two, Units 3 and 4 at the Vogtle plant in Georgia, were finished as a result. By the time they were finally turned on, in 2023 and 2024, they were years behind schedule and $20 billion over budget. A similar attempt to build new reactors at the VC Summer plant in South Carolina ended with a very large hole in the ground and local ratepayers on the hook to pay it off.
In the wake of Three Mile Island, scholars and government officials ascribed the nuclear industry’s malaise to various causes: regulation of a young industry, inept management and design, a corrupt nexus between the nuclear sector and political power. Plant operators were meanwhile focused on safety, not only for the sake of their plants but because Three Mile Island had underlined that the public’s trust was a shared resource. “Each licensee a hostage of every other licensee,” was how the CEO of Detroit Edison put it in 1989.
In the past decade or so, though, it has become more common to see arguments that lay the blame at the foot of the NRC. Take, for example, “It’s the Regulation, Stupid,” a 2024 essay from Ted Nordhaus of the Breakthrough Institute. Or listen to the Democratic governor of New York, Kathy Hochul, calling for the construction of an advanced nuclear plant in the state last June: “There’s a reason people don’t embrace nuclear energy—a lot of reasons. One of them: It just takes too long, and the barriers are in Washington.” Hochul went on to describe how, in a conversation with the president, she encouraged Trump to have “Elon Musk and all the ‘DOGE guys’ go over to the Nuclear Regulatory Commission.”
Taylor shares the deregulatory impulse that lately goes under the slogan of abundance. His lawsuit against the NRC originated with the Abundance Institute and a former Chicago University law professor who, with financial support from the Koch brothers, had created an investment firm dedicated to “regulatory entrepreneurship.” And he has tweeted favorably about Ezra Klein and Derek Thompson’s book, Abundance, lamenting that Democrats would have to “fully exorcize the climate-communism demon” before they could embrace such a platform.
The bedrock of all this is his conviction that he should be able to build a reactor and test it without significant interference from the government. “In aviation, we have a strong independent regulator, but we also have a robust testing framework,” Taylor said. “So you can go and you can fly an airplane out in the desert and you can even crash it. And, you know, you try not to crash it, but if you do, it doesn’t blow up the company, right?” There was a place for a regulator, he allowed, but it came well after testing had been allowed to flourish. I asked if Valar had had any engagement with the NRC. Taylor said it hadn’t.
3.
“I’ve said to people, an awful lot of what’s currently happening at the NRC feels like an Oklo revenge tour,” one former government official with knowledge of these events said to me. In 2020, Oklo Inc. was the first company to apply to the NRC for a construction permit to build an advanced reactor, or one that is not cooled with water. After two years of acrimonious back and forth, during which Oklo’s application never moved beyond the preliminary review, the NRC sent the company a letter informing it that its application had been rejected. The agency cited Oklo’s failure to provide “detailed technical information responsive to the staff’s requests for details about the safety of [Oklo’s] design.” Oklo’s CEO, Jacob DeWitte, has accused the NRC of screwing up. The executive orders that Trump signed on May 23 last year took Oklo’s side. “Instead of efficiently promoting safe, abundant nuclear energy, the NRC has instead tried to insulate Americans from the most remote risks without appropriate regard for the severe domestic and geopolitical costs of such risk aversion,” reads the second of the four. The same order goes on to call for a “wholesale revision” of the NRC.
That process began only weeks later, when actual DOGE guys appeared at the NRC’s headquarters in June. Adam Blake, an investor and former CEO of various energy and AI companies, was detailed to the NRC as the agency’s new DOGE lead. He arrived with a retinue of other lawyers and staff, kicking off a cascade of personnel and institutional changes at the NRC that effectively folded the agency under the authority of the DOE and the White House. Part of that work had already been accomplished under an executive order from February, in which the president had claimed a right of approval over the agency’s rulemaking. Beginning in June, DOGE staff and the president also began implementing more direct forms of control. On the 16th, Trump fired Christopher Hanson, a Democratic appointee and the former chair of the NRC’s five-person commission. A steering committee was then stood up and staffed with DOGE affiliates to implement Trump’s executive orders, including the rewriting of the agency’s rules.
So far, their recommendations have suggested changing environmental-impact reviews, cutting the number of inspections for operating plants, allowing nuclear workers to sustain higher doses of radiation, and sunsetting the NRC’s aircraft impact assessment, which requires nuclear power plants to demonstrate that a large plane crashing into the reactor would not produce to a major release of radioactivity. Many of these reforms have been long-standing desiderata of the nuclear industry, and the last few years have seen several legislative efforts to speed up NRC licensing, most recently with the ADVANCE Act, passed during Biden’s last year in office. Under Trump, they are being carried out by fiat. In a recent ProPublica article, a young DOE lawyer who had entered government through DOGE, Seth Cohen, is reported to have commented during an internal meeting: “Assume the NRC is going to do whatever we tell the NRC to do.”
I talked with a number of people who expressed profound concerns about these developments. Scott Morris was number two at the NRC when he retired in May 2025, the same week that Blake, Cohen, and other DOGE staff first turned up at the agency. Morris voted for Trump twice, and he is personal friends with Trump’s choice for the current NRC chair, Ho Nieh, who used to work for Morris when he was a senior resident inspector in the 1990s. Even so, Morris has lately been outspoken in his concerns about what has gone on at the NRC. “I understand the desire for innovation. I support that. I desire efficient regulation,” Morris said. But he insisted that reforms needed to be done the right way. “It’s one thing to challenge the status quo and try to innovate. It’s another to try to go behind closed doors and blow the whole thing up.” Probably the thing that bothered Morris the most was the lack of transparency in the new dispensation. At the NRC, the license applications and all the correspondence between the regulator and the applicant are made available online. In the new pathways created through the DOE, there are no plans to make the regulatory work public. “I’m a government-in-the-sunshine guy,” he said. “I’m a prove-it-to-me guy. I’m not just going to take your freaking word for it.”
The NRC also came up when I talked with Doug Robison, a third-generation “oil and gas man” who founded a nuclear energy company called Natura Resources in lieu of retirement. Like Valar, Natura is part of the DOE’s Reactor Pilot Program. Yet Natura has the distinction of being the only company in the pilot program that already has a construction permit from the NRC. (It is one of only three nuclear companies in the country able to make that claim.) “Two years to prepare that document right there,” Robison said, pointing me to a leather-bound book on a side table in his office. “Eighty examiners, 13,000 man-hours of review, answering over 300 questions by the NRC, all really pointed at: Is this safe to deploy?” The licensing process had taken four years all told. For Robison, it was a badge of honor. Logically, he said, he could understand the desire to go around the NRC, but he didn’t agree with it. “Scares the hell out of me to think about deploying a reactor that hasn’t been through the crucible that the NRC puts us through. And I almost feel like if you don’t have that thought in your brain, maybe you shouldn’t be deploying nuclear reactors.”
4.
For Valar, the new look at the NRC was an enormous benefit. Before Trump added his signature to the executive orders, Valar employees were exploring land in southern Utah for a place to put the company’s first gigasite. (Like many people with a connection to the nuclear industry, they had known what was coming.) Jaron Wallace, the director of the Utah San Rafael Energy Lab, said he crossed paths with them at this time. That lab opened in 2018 in the heart of Utah’s coal country with plans to host a local nuclear energy startup, but the startup went bankrupt. With 40 acres of mostly unused land and offices in a converted warehouse, Wallace was looking for projects. He explained this to Valar. “Between when Valar told us they wanted to build a reactor here and a signed contract was about two weeks,” Wallace told me.
Securing a plot at San Rafael let Taylor announce plans to build a test reactor on the same day that the executive orders were announced. From there, things just kept falling into place for him and his company. In August, Valar was selected as one of 10 companies to take part in the DOE’s Reactor Pilot Program. That gave it preference for fuel allotment and a fast track to regulatory approval for its test reactor through the DOE.
All companies in the pilot program benefited from the same structure, but Valar appears to have enjoyed a particularly close relationship with the former DOGE staffers who were spearheading reforms at the NRC. Valar has company hats that read “Make Nuclear Great Again,” written and laid out in the style we all are familiar with. Seth Cohen is reported to have handed out these hats to NRC staff at one meeting. In other instances, Valar’s expenses were weighed in discussions of where to set the radiation-exposure limits for nuclear workers, and when someone raised concerns about putting an experimental reactor near a population in Utah, a member of the group cracked a joke, sort of, that those people had “been downwind before.” This was a callback to how rural populations in southern Utah had been exposed to heavy radiation doses during the nuclear weapons testing in the 1950s and ’60s.
The real lift for Valar came in November, however, with a Series A funding round led by Snowpoint Ventures, Dream Ventures, and Day One Ventures. (Snowpoint is a major firm founded by a former head of global defense at Palantir. Dream Ventures is a bit of a cypher; it has a website with a logo in one corner and the words “Investing in Extraordinary Dreamers” displayed prominently, with no other information. Day One Ventures was founded by Masha Drokova, an émigré who was a high-ranking member of Russia’s nationalist youth movement, Nashi, before becoming disenchanted with Vladimir Putin. In the States, she got her start in venture capital while working as Jeffrey Epstein’s publicist from 2017 to 2019. When I asked Valar’s director of communication about Drokova, I was told that she’s not on the board.) The funding round brought in $130 million, much of it from Shyam Sankar, Palantir’s chief technology officer and executive vice president, as well as from Palmer Luckey, the founder and head of the defense company Anduril Industries. (I wrote to both of them asking to speak about their choice to invest in Valar and received a polite no from each.) With that money, Valar had more than enough to build its experimental reactor in Utah. As a first step, it brought its reactor core critical at Los Alamos. Taylor claimed that Valar was the first startup to “split the atom,” rowing that back after it was pointed out that other venture-backed companies had done it years earlier.
Work at the San Rafael Energy Lab moved quickly. Valar blasted the ground for its reactor in September. Workers poured the concrete pad over the following months and threw up a large hangar emblazoned with Valar’s branding across the gable. Then in February, with cameras and press in tow, they loaded a reactor vessel into a C-17 Globemaster III aircraft supplied by the U.S. military and flew it from Southern California to Utah. At the press conference, held at Utah’s Hill Air Force Base, where the jet touched down, Taylor was wearing a pair of aviators and a flight jacket, and he reiterated Valar’s commitment to hitting the July 4 deadline. After more speeches, Energy Secretary Chris Wright took the stage. Wright was a member of the board at Oklo before he assumed public office, and his comments were trained on regulation. America had built more than a hundred reactors, “relatively rapidly,” in the decades after the Second World War. “Then we let bureaucracy get in the way. We created the Nuclear Regulatory Commission over 40 years ago, and since then we’ve permitted, approved, and built two nuclear reactors. We stopped an emerging technology dead in its tracks.”
I noticed on the day this flight took place that Nick Touran, the engineer who confronted Taylor over his claims about holding spent fuel, had tweeted fairly positively on the news that Valar’s reactor had taken flight. (“Love seeing this badass progress after so long,” he wrote, before pointing out that other reactor vessels had also been put on planes before.) When I spoke with Touran, he told me that he had changed his views on Valar a bit. “They sort of really value getting things right, and so they’ve updated the calcs and understanding and now they’re not gonna hold the fuel when it comes out of the reactor,” he said. It wasn’t that he thought Valar was going to necessarily succeed. Touran expected Valar’s reactor to be under-engineered, like any others that hit the July 4 deadline. “They’re probably gonna break down all the time, and they’re gonna have all sorts of maintenance,” Touran said. But the venture itself, movement in an industry that had been frozen for decades, was worth supporting.
I was able to speak to Taylor again around this time. The day Valar had flown the reactor vessel to Utah was one of the coolest days in his life, he said, and he was talking about Valar becoming “a trillion-dollar company.” I asked what it was like to work with the DOE as a regulator. He told me the DOE applied the same standards in its safety assessment as the NRC would have, except that the DOE was much more engaged. “The DOE will have a conversation with you,” he said. “It’s very, very collaborative to meet the standard. The NRC is more like: come completely ready. And there’s very little dialogue.” That aligned more or less with what I had heard from Rian Bahran, the deputy assistant secretary of nuclear reactors at the DOE and the person in charge of the Pilot Program. (“The culture is give me what you have. Let’s do table reads. Let’s iterate together,” Bahran had said.) Nevertheless, I was a bit surprised when I heard Taylor say that Valar was planning to submit an application to the NRC. “Once we’re ready,” he said. “We’ve met all those standards. We have all of the data. We have the empirical testing. That’s the right time to step into the NRC court.”
Valar’s suit against the regulator had long been backburnered. On April 30, both parties filed a joint motion for a 60-day stay. When one stay expires, they file again. The case remains in abeyance today. Earlier in April, according to reporting from Bloomberg Businessweek, one of the plaintiff’s lawyers emailed his colleagues to tell them that the NRC was willing to settle, ceding regulatory authority over some small reactors to state agencies. Utah is already moving to set one up. In January this year, one of the lawyers who represented the corporate plaintiffs on the suit, Trent McCotter, withdrew from the case to join the Office of the Deputy Attorney General at the DOJ.
5.
San Rafael Energy Lab sits just outside the small town of Orangeville, in the foothills of a chain of high mesas. I arrived early on the day I was supposed to receive a tour of Valar’s reactor from Taylor. To kill time, I drove by the PacifiCorp coal plant just outside of town and watched its smokestacks quietly exhausting into the sky. Trump had started a war with Iran about a month earlier, and the price of gas and plane tickets was climbing quickly. That crisis overlaid what looked like an impending drought. Southern Utah had broken heat records that March, and the heat had vaporized almost all of what was already the lowest snowpack in the state’s record. Salt leaked from the ground in a dried-out arroyo I could see from the road.
About an hour before my tour, Tara Harandi-Zadeh, an early investor in Valar who has been running its comms, called me to say that Taylor had missed his flight in D.C. and wouldn’t be able to make it. That was a little strange. Taylor had to have missed his flight right around when I had confirmed with her that morning, before I started the two-and-a-half-hour drive down from Salt Lake. I said as much when I met Harandi-Zadeh in a corrugated steel warehouse on San Rafael’s campus. She talked about a backup plan to fly Taylor directly to the site that had also fallen through, but the gist of the story was that he hadn’t told her until shortly before she called me. That seemed plausible enough, given that she traveled to southern Utah from London to help orchestrate the tour, and that the two Valar employees she had assigned to show me around seemed to have been as nonplussed as I was.
Sean Stassi was an incredibly good-natured 21-year-old and one of Valar’s mechanical engineers. He had met Taylor at a bonfire on the beach at El Segundo three years earlier. My other tour guide was Jess Housekeeper, a local approaching 40, with slate-blue eyes and a long history working in heavy industry, including a stint as an instructor in the Navy’s nuke school. I recognized Housekeeper from an audio recording of a local town hall Valar held in July 2025. He had been working at the PacifiCorp coal plant then, and during the public comments, he had stood up to express his strong support for nuclear energy in the county. Valar hired him a few months later as director of Utah operations.
After Housekeeper gave me a hi-vis yellow vest and the other necessary PPE, I followed him and Stassi past a chain-link fence that separated Valar’s reactor building from the rest of the San Rafael lab. Inside the fence was an active construction site, with loaders and cranes moving around at the periphery and a man in a hard hat working away on some task seemingly every couple of yards. “There’s big stuff that can run you over, so just stay next to us if you could,” Housekeeper said. As we walked, Stassi told me about the Airbnb where he was staying, which kept zebras in the field.
Our first stop was a bunch of large concrete blocks. “A lot of people ask us, How do you go so fast and stay safe?” Housekeeper said. The blocks were an example. They had been cast in Salt Lake and brought down to the site where they would be assembled into a radiation shield for the core, once Valar got the DOE’s sign-off to put fuel in it.
Each block was shaped so that, along one axis, it bent kind of like an accordion. Stassi had come up with the design, and it seemed fairly ingenious. When the blocks were stacked one atop the other, he explained, the shape ensured that each block would fit into the others like teeth, denying radiation a straight-line gap through which it might pass.
I asked if these blocks had been qualified yet—that is, whether they had been tested under high radiation to show that they would function as expected. “We have shown in our simulation that we want a certain density to stop gamma. We have material test reports showing that all these blocks adhere to that density. And then we’ll have live radiation monitoring on the plant to verify that,” Stassi said. What I took from his answer was that the first live test would come when Valar fired up the reactor.
From the shielding we continued to a small prefab building, inside of which was the control room. Outside we had been in a world of wiring, dusty light, and construction workers, but the control room was maybe the most finished space on the site that I saw. Done up in matte black panels and frosted strips of faint purple light, it looked like an interior space from the Death Star. Housekeeper directed my attention to the middle of the control panel. “So, as they say, here’s the big red button,” he said. Pressing it would send six control rods into the core, soaking up enough of the free neutrons to stop the fission process and scramming the reactor. The rods were held up electrically, he continued, so that if the power cut, they would drop in automatically.
One of the big selling points of high-temperature gas reactors like the one Valar might soon be turning on is that, even if the active safety systems fail and the control rods are not put into the core, the properties of the reactor are such that it can theoretically bleed off enough heat fast enough to avoid a meltdown. This is pretty common with new reactor models and is billed as passive safety. I asked about it. “Oh yeah, we could go way into that,” Housekeeper said. “Yeah, like that’s our shit,” Stassi agreed, and they proceeded to tell me about their reactor cavity cooling system. Besides the very small output of Valar’s test reactor, which the company plans to bring up to 250 kilowatts of power before dropping it back into the hundreds, the other significant piece of Valar’s safety case is its choice of fuel. Called TRISO (for “tristructural isotopic”), the fuel is fabricated so that every uranium particle is encased in a ceramic coating that can withstand extremely high heat and will contain within it nearly all the radioactive fission products that are created as the uranium starts splitting. There is still plenty of radiation to be worried about, in the form of waves and neutrons, but TRISO dramatically reduces the chances of a major disaster. The big downsides are that TRISO is expensive to make, and there is very little available. Valar was planning to manufacture its own on-site, but that facility was nothing more than a patch of concrete when I saw it.
Finally, we entered the reactor building. A large U.S. flag had been stuck to the wall, and the ground was a vast pad of exposed concrete that ran several feet deep. Near the center of this pad, looking somewhat small within the hangar’s voluminous interior, the reactor vessel stood upright, a rounded steel cylinder maybe 15 feet high and painted black. In Valar’s design, helium will draw the heat off the reactor core through a U-shaped pipe that runs through a trench and up again into an Escheresque complex of what looked like off-the-shelf steel ducts. These contained a heat exchanger, a purification system for the helium, and a squat red vessel, studded with steel bolts, that will pump the helium through the system.
“It’s beautiful,” Housekeeper commented. “It’s a piece of art, like a lot of engineering.”
Alvin Weinberg, one of the founding figures of civil nuclear energy in the United States, once described a reactor he was building as “a pot, a pump, and a pipe.” That undersells things a bit, given that the fission process unleashes some of the most awesome physical forces in the world. But it’s a pretty good description of what reactors look like. You can’t actually see the fission process. The rest is mostly plumbing. Even so, I let myself lean into the industrial chic of the apparatus before us and could agree with Housekeeper that there was something Promethean and sort of incredible about what we were looking at. Only a few months earlier, none of this had existed. It had been conjured on the spot by human labor and tens of millions of dollars.
There was still obviously a lot left to do. On our way back to Valar’s on-site office, we stopped in the transition room, where Valar’s nuclear operators will prepare themselves before and after they enter the reactor building. There was an emergency sink to wash out your eyes and a naked shower head serving a similar purpose. Otherwise, it was close to unfinished. Next to the door that separated this room from the reactor, there was a massive hole in the wall where I could see naked metal studs and a tuft of pink insulation poking out from the drywall. Housekeeper stationed himself in front of that hole and pantomimed a safety check. “You’d set your feet and hands inside a monitor, and it would scan you,” he said. “And it gives you a green light and tells you you’re clean and then you would walk over here and you know you’re clean going out.” I had been snapping pictures throughout the tour, and I lifted my phone to take one here.
“You wanted a picture of the hole in the wall?” Housekeeper asked.
“Yeah,” I said. There was a slightly uncomfortable moment we passed over with laughter. “But it’s coming?” I said, meaning the monitor Housekeeper had described. They assured me it was.
Throughout the tour, Housekeeper in particular had been keen to impart the paramountcy of safety protocol. “We’re just here to keep our personnel, the public, and the environment safe,” he told me. But clearly they weren’t ready to do that yet. They said they would be, and it would be up to the DOE to decide. As far as I know, no one but the DOE and Valar will be able to see what that decision was based on.
We returned to Valar’s office in the warehouse and talked a bit more. Housekeeper told me a bit about how important electricity was to the well-being of people around the world and suggested a YouTube documentary on energy I should watch. I asked if he had seen the ProPublica article where someone from DOGE had joked about local people being downwinders. He hadn’t read it, he said, and didn’t want to comment without doing so. After a little more talk, Housekeeper said he needed to get to the airport and made his apologies. Leading me around on an unplanned tour had kept him longer than he’d intended. He had been gracious about it.
With Housekeeper gone, I continued to talk with Stassi and Harandi-Zadeh, Valar’s acting comms contact. I posed a question I had been thinking about for a while. Were they worried that Valar’s close association with the Trump administration might start to play negatively for them? Harandi-Zadeh pointed out that Oklo’s CEO had been standing in the Oval Office when Trump signed the executive orders. “Surely they were like way closer,” she said. We went back and forth on this point a bit, and I brought up Valar’s “Make Nuclear Great Again” hats.
Here Stassi weighed in. “I mean, they’re funny, right? Like it is a ridiculous thing,” he said. “Even my mom, who’s never voted Republican in her life, has one and thinks it’s kinda funny.” Before I left, he showed me a whiteboard on the office fridge with standing orders for snacks and the like. ZYN nicotine pouches had been written on it three times, along with chocolate milk, kombucha, and creatine.
6.
The audio recording of the town hall that Valar held in July 2025 is a fascinating document. This was the same town hall where Jess Housekeeper had voiced his support for nuclear energy. By that time, Valar had its site picked out and construction was moving forward. In a squat brick building that holds the Emery County Office, Taylor and other senior executives faced people who had had very little say in any of it.
The issue of safety came up repeatedly throughout the night. A local woman asked whether the reactor had run before. “We had it on since February. On and off, in different ways,” said Max Ukropina, Valar’s head of projects and a former GOP congressional candidate for Orange County, California. It became clear that Ukropina was talking about Valar’s thermal prototype and not the actual reactor the company was bringing to the county. “So this technically is untested,” the woman correctly concluded.
Another questioner brought up Valar’s close association with Trump in an accusatory way, and a local rancher read out part of an article he had found on Google in which a professor of radiology had criticized Valar for not submitting its design to the NRC. At a few points, the questioners got heated, but for the most part the conversation stayed calm. There were almost as many questions about the jobs and economic benefits Valar’s reactor might bring as there were about its hazards.
Taylor and his executives remained cordial in answering, and Taylor in particular was careful not to imply that Valar’s reactor would mean an economic windfall in the near term. (“We’re hesitant to be the company that comes in and promises the world and then is unable to deliver, right?”)
Just past the two-hour mark, a local resident who said he had five daughters brought up the Chernobyl disaster for the fourth time that evening, by my count. He pressed Taylor on the idea that his town was under the obligation to service a geopolitical energy race. “We don’t want to compete against China. We want our small town here,” he said. “We don’t want to destroy it. I don’t want to destroy it for my daughters either.”
“Do you mind if I ask you a question?” Taylor said. “How many people in the two counties we’re talking about here, Carbon County and Emery County, how many people have died in the coal industry?”
“Hundreds. Same way back home in Pennsylvania,” said the man. “We couldn’t even fly-fish when I was a kid because it was so polluted. We can now.”
“And do you think that that’s been generally worth it?” asked Taylor. “You know, the deaths are terrible, but power is important.”
