⚡ Energy
It Took Nine Months to Build a Nuclear Reactor from an Empty Lot. The NRC Would Need That Long Just to Review the Paperwork.
Valar Atomics’ Ward 250 achieved zero-power criticality on June 18 at the Utah San Rafael Energy Lab, nine months after the site was bare dirt. It is the second advanced reactor to go critical under Executive Order 14301, the first built outside a national laboratory, and a live demonstration that the DOE’s fast-track pathway can compress a decade of nuclear development into three quarters of a year. In the time it took to finish Vogtle 3&4, this pathway could have cycled through 16 reactor designs.
Nine months.
That is how long it took to go from an empty patch of Utah desert to a self-sustaining nuclear chain reaction. On June 18, Valar Atomics’ Ward 250 reactor achieved zero-power fueled criticality at the Utah San Rafael Energy Lab in Emery County, confirmed by the Department of Energy’s Office of Nuclear Energy at roughly 4:30 p.m. MDT. A convoy of three C-17 Globemaster III aircraft had carried the reactor’s eight modules from March Air Force Base in California to Hill Air Force Base in Utah just four months earlier. “Nine months ago, this was an empty site,” CEO Isaiah Taylor said. “Today, there’s a critical reactor on it, built and operated by the Valar team.”
Ward 250 is a Gen IV high-temperature gas reactor that uses TRISO fuel (uranium kernels encased in ceramic layers), helium coolant, and graphite moderators. At 100 kilowatts thermal for the test configuration, it is small. But the milestone it represents is not. It is the second advanced reactor to achieve criticality under President Trump’s Executive Order 14301, which directed the DOE to hit three by July 4, 2026. It is the first DOE-authorized reactor ever built and operated outside the national laboratory system. And it has already begun non-commercial power ascension, with further milestones expected in the coming weeks.
Two down, one to go, and twelve days to hit three.
A Race Nobody Thought Would Be Close
When Trump signed EO 14301 in May 2025, calling for at least three advanced reactors to reach criticality by Independence Day 2026, the target struck most nuclear engineers as aspirational bordering on fictional. Getting a single reactor from concept to criticality typically consumes a decade or more of regulatory review, design iteration, fuel qualification, and construction. Getting three in 13 months, from ten companies that ranged from well-funded startups to firms most people in the industry had never heard of, seemed like a political gesture dressed up as a technical goal.
It wasn’t.
Antares Nuclear’s Mark-0 went critical at Idaho National Laboratory on June 4, becoming the first privately developed non-light-water reactor to reach that milestone in the US in over four decades. Fourteen days later, Valar matched it from a standing start in the Utah desert, with no national lab infrastructure to lean on. Now Aalo Atomics expects to hit criticality with its 10-megawatt-electric Aalo-X sodium-cooled reactor at Idaho National Laboratory “well before” the July 4 deadline. All three cleared design reviews, signed DOE Other Transaction Agreements, sourced fuel, and built reactors within approximately a year of selection.
For comparison: NuScale Power spent 23 years developing its small modular reactor before securing NRC design certification in 2023, then watched its flagship Idaho project implode when costs spiraled to $9.3 billion. TerraPower, founded by Bill Gates in 2008, broke ground on its Natrium reactor in 2024 and doesn’t expect operations until roughly 2030. The Romania deployment that will carry NuScale’s first commercial modules to a grid won’t begin operations until 2033, twelve years after project start. In September 2025, ENTRA1 Energy and the Tennessee Valley Authority agreed on a 6-gigawatt NuScale SMR deployment, the largest in US history, but that program won’t deliver its first watt for years.
The Regulatory Speed Gap
Speed is not an accident here; it is a design choice baked into the DOE Reactor Pilot Program’s legal structure.
Under the traditional pathway, the Nuclear Regulatory Commission reviews reactor designs, issues construction permits, oversees building, and licenses operation, with each phase consuming years. Nano Nuclear Energy submitted a Construction Permit Application for its KRONOS microreactor at the University of Illinois in March 2026. The NRC formally accepted it in April, and Nano Nuclear estimates the technical review will take approximately 12 months, after which a construction permit could arrive in the second half of 2027. Only after that can on-site nuclear construction begin. And KRONOS is a 15-megawatt-electric gas-cooled reactor not all that different in concept from Ward 250: same TRISO fuel, same helium coolant, same basic physics, but wildly different timelines.
By the time the NRC finishes reviewing KRONOS’s paperwork, Valar will have been operating a critical reactor for over a year.
The DOE pathway works differently because Executive Order 14301 gave DOE authority to authorize reactor testing directly, bypassing NRC jurisdiction for pilot-stage demonstrations at DOE sites and approved off-site locations. Companies submit preliminary design safety analyses and documented safety analyses to DOE, which reviews them on a compressed schedule. Valar’s preliminary design analysis was approved in February 2026, its documented safety analysis in late April. Construction, fueling, and criticality all happened under DOE oversight on a timeline that a conventional NRC process would need just to hold its first public hearing.
This creates a measurable speed gap, and I counted the months:
| Milestone | DOE Pathway (Valar) | NRC Pathway (Nano Nuclear) |
|---|---|---|
| Application to permit/authorization | ~5 months | ~16 months (est.) |
| Permit to criticality | ~4 months | ~18–24 months (est.) |
| Total: concept to criticality | ~9 months | ~34–40 months (est.) |
| Speed ratio | ~4:1 | |
Four to one. And that comparison flatters the NRC pathway, because Nano Nuclear’s KRONOS prototype is a research reactor at a university, not a commercial plant. A full commercial license adds years on top.
Sixteen Shots vs. One
Here is the calculation that reframes what this means for the 400-gigawatt target.
Trump’s executive orders call for expanding US nuclear capacity from roughly 100 GW today to 400 GW by 2050. That is 300 GW of new capacity in 24 years. Even optimists concede that individual microreactors producing 5 megawatts each cannot close that gap on their own; you would need 60,000 of them. Commercial-scale plants like NuScale’s 924-megawatt twelve-module configurations are designed for that job, but they need a decade per plant under current NRC processes, and 325 of them to reach the target.
So the real value of a 9-month criticality cycle is not in the kilowatts it produces. It is in the design iterations it enables.
Vogtle Units 3 and 4, the only new nuclear reactors completed in the US in a generation, took roughly 12 years from construction start to commercial operation and cost approximately $35 billion for 2.2 gigawatts. In that same 12-year window, the DOE fast-track pathway could theoretically support 16 complete reactor design-build-test cycles. Sixteen chances to learn what works, fail fast on what doesn’t, and feed proven designs into the NRC commercial licensing pipeline with actual operating data instead of theoretical safety analyses. Silicon Valley ships products every sprint and learns from each one. Before EO 14301, the nuclear industry shipped a product every generation and learned from almost nothing, because almost nothing got built.
Valar is already moving to capitalize on this. Ward 250 has begun non-commercial power ascension, and the company’s stated goal is power generation by July 4. Its commercial vision involves building “nuclear gigasites,” clusters of thousands of HTGRs designed to produce industrial power and carbon-based fuels. Whether that specific vision materializes or not, the operating data from a 9-month test cycle is infinitely more useful than another decade of paper studies.
The Ten-Company Scorecard
The DOE selected ten companies for eleven projects in the Reactor Pilot Program. Here is where they stand with 12 days until the deadline:
| Company | Reactor | Status |
|---|---|---|
| Antares Nuclear | Mark-0 (HTGR) | ✅ Critical (June 4, INL) |
| Valar Atomics | Ward 250 (HTGR) | ✅ Critical (June 18, Utah) |
| Aalo Atomics | Aalo-X (Na-cooled, 10 MWe) | CTR unveiled at INL; criticality expected before July 4 |
| Oklo | Aurora (fast reactor) | Construction underway (started Oct 2025) |
| Deep Fission | Deep borehole reactor | Groundbreaking in Parsons, KS (Dec 2025) |
| Last Energy | PWR (20 MWe) | Construction underway in Texas |
| Radiant Industries | Kaleidos (1.2 MWe) | DOE authorizations in progress |
| Natura Resources | Natrium variant | DOE authorizations in progress |
| Terrestrial Energy | IMSR (molten salt) | DOE authorizations in progress |
| Atomic Alchemy | Isotope production reactor | DOE authorizations in progress |
Three companies are in the hunt for the July 4 deadline, four more are under active construction, and three are still navigating DOE permitting stages. That distribution matters: it demonstrates that even an aggressive fast-track program cannot compress every design into a one-year timeline. Some reactor concepts are further from ready, and the program correctly lets them move at their own pace rather than cutting corners.
Strongest Counterargument
The obvious objection is safety. The NRC exists because nuclear accidents have consequences that dwarf those of a failed software sprint, and its deliberate pace reflects a regulatory philosophy forged in the aftermath of Three Mile Island and refined after Fukushima. Bypassing NRC jurisdiction for DOE-authorized test reactors may be appropriate for small-scale demonstrations with walk-away-safe designs using meltdown-resistant TRISO fuel, but that logic does not automatically extend to commercial fleets generating gigawatts. If the DOE fast-track pathway produces a design that later enters NRC commercial licensing and something goes wrong during that transition, the political backlash could set the entire nuclear renaissance back further than any regulatory delay ever did. Every month the NRC “wastes” on review is also a month spent building a safety case that protects both the public and the industry from a catastrophe that would end both.
Limitations
Ward 250 is a 100-kilowatt-thermal test reactor, not a power plant. Zero-power criticality means a self-sustaining chain reaction at essentially no measurable energy output. All EO 14301 reactors are demonstration units; none have generated commercial electricity. The 4:1 speed ratio in the table above uses estimated NRC timelines for Nano Nuclear’s KRONOS, which could come in faster or slower than projected. The 16-iteration calculation is theoretical and assumes each cycle involves a genuinely different design, not repetitive builds of the same reactor. HALEU fuel supply remains constrained, with total US production insufficient to support a commercial microreactor fleet, as our earlier reporting on the Antares milestone detailed. And the DOE pathway’s legal authority for off-site reactors is an executive order, which a future administration could revoke.
The Bottom Line
For forty years, the US built zero advanced reactors. In June 2026, it built two in two weeks, and a third is on deck before the fireworks. If the only thing that changed were the reactors themselves, that would be a technical achievement worth noting and nothing more, but what actually changed is the pathway. Executive Order 14301 created a parallel regulatory track that delivers criticality demonstrations four times faster than the NRC process can deliver a construction permit review. For anyone tracking the 300-gigawatt gap between where US nuclear capacity is and where it needs to be, the bottleneck just shifted from physics to paperwork. If you work in energy policy, utilities, or nuclear supply chains, the action items are now: study the DOE Reactor Pilot Program’s authorization structure, identify which commercial reactor designs will emerge from it with operating data in the next 2–3 years, and prepare to meet them at the NRC door with de-risked licensing applications. If you are an investor, watch Aalo Atomics in the next 12 days, the DOE’s fuel line pilot selections for TRISO and HALEU supply, and whether Congress codifies the fast-track pathway into statute before a future executive order can undo it.