For the first time in more than eight years, the United States has authorized construction of a new commercial nuclear power plant. And it did not stop at one.
In a span of months, the Nuclear Regulatory Commission granted construction permits for two advanced reactor projects: TerraPower’s Natrium plant in Kemmerer, Wyoming, and Kairos Power’s Hermes 2 test reactor near Oak Ridge, Tennessee. The back-to-back approvals break a permitting drought that stretched back to the combined operating licenses for the Vogtle expansion in Georgia, and they mark the first time the NRC has cleared construction of reactor designs fundamentally different from the light-water technology that has dominated U.S. nuclear power for half a century.
The Kemmerer permit: a commercial first
The higher-profile milestone belongs to TerraPower, the company founded by Bill Gates in 2008. In early 2025, the NRC issued construction permit CPAR-1 to US SFR Owner, LLC, a TerraPower subsidiary, for Kemmerer Power Station Unit 1. The Associated Press reported it as the first commercial nuclear construction permit the agency had granted in more than eight years.
The Natrium design pairs a sodium-cooled fast reactor with a molten-salt energy storage system. At its base, the reactor generates roughly 345 megawatts of electricity. When the storage system discharges, output can surge to approximately 500 MWe, according to the Department of Energy’s environmental assessment EA-2264 for preconstruction funding. That flexibility matters: unlike traditional nuclear plants that run at a fixed output around the clock, the Natrium plant is designed to ramp up when the grid needs more power and pull back when wind and solar are producing heavily. Grid operators have been asking for exactly that kind of dispatchable generation as renewable capacity grows.
The site itself was chosen with economics in mind. Kemmerer sits next to a retiring coal plant, and TerraPower plans to reuse existing transmission lines and grid connections. Securing new transmission access can add years and hundreds of millions of dollars to a power project, so plugging into infrastructure that already exists gives the plant a meaningful head start.
The federal environmental review followed a defined path. A Notice of Intent was filed on June 12, 2024, and the Final Environmental Impact Statement, cataloged as NUREG-2268, reached the Federal Register on November 28, 2025. The EPA’s own EIS database entry confirms those dates and shows the NRC completed a full environmental analysis rather than a narrower assessment. Separately, the DOE’s EA-2264 covered federally funded preconstruction work, meaning two parallel regulatory tracks had to converge before construction could begin.
Hermes 2: a test reactor with commercial ambitions
The second project moved through the NRC even faster. On November 21, 2024, the commission granted construction permits to Kairos Power for the Hermes 2 test reactor facility, documented under ADAMS accession numbers ML24324A021 and ML24324A022. Hermes 2 is classified as a non-power test reactor, not a commercial plant, but its approval still required full NRC safety and environmental review.
Kairos uses a molten-fluoride-salt coolant and ceramic-coated fuel pebbles, a combination the company argues can be manufactured more cheaply and built more quickly than conventional reactor systems. The Hermes 2 facility is meant to prove that theory under real operating conditions. Whether the results feed directly into a commercial-scale Kairos reactor or require additional licensing rounds is not spelled out in public filings. For now, the permit is best understood as a bet on learning: valuable, but not yet a source of grid electricity.
The NRC’s own 2025 advanced reactor chronology lists both the Kairos and TerraPower actions alongside progress on a third project, the Tennessee Valley Authority’s Clinch River site, where a construction permit application for GE Hitachi’s BWRX-300 small modular reactor is under active review. A permit application in progress is not the same as a permit granted, and that gap has historically stretched for years. But the fact that three advanced designs are moving through the NRC pipeline simultaneously is itself unusual.
Why the drought lasted so long
To understand why two permits in quick succession qualify as news, it helps to look at what came before. The last time the NRC authorized commercial nuclear construction was in 2012, when it issued combined operating licenses for Vogtle Units 3 and 4 in Georgia. Those reactors, based on Westinghouse’s AP1000 design, became a cautionary tale: years of delays, billions of dollars in cost overruns, and the bankruptcy of the lead contractor. Vogtle Unit 3 finally entered commercial service in 2023 and Unit 4 in 2024, but the experience left utilities, investors, and regulators deeply wary of new nuclear builds.
The 2011 Fukushima disaster in Japan had already chilled global appetite for nuclear expansion. In the U.S., cheap natural gas from the shale revolution made new reactors even harder to justify on cost alone. Several existing plants shut down because they could not compete with gas-fired generation, let alone fund new construction.
What changed? Several forces converged. The Inflation Reduction Act of 2022 created production tax credits for zero-carbon electricity, including nuclear. The ADVANCE Act, signed into law in July 2024, streamlined NRC licensing fees and review timelines for advanced reactor designs. And a wave of power-purchase agreements from major technology companies, driven by the enormous electricity demands of data centers and artificial intelligence, injected private capital and urgency into the nuclear pipeline. Microsoft, Amazon, and Google have all struck deals tied to nuclear generation in the past two years, signaling that deep-pocketed buyers see reactors as part of the answer to their growing energy needs.
What the permits do not guarantee
A construction permit is a legal authorization to pour safety-related concrete and erect major structures. It is not a guarantee that a plant will be finished on time, within budget, or at all.
Neither the NRC filings nor DOE documents in the public record specify a firm completion date or total capital cost for Kemmerer Power Station Unit 1. Without those figures, it is difficult to judge whether the project can compete on price with natural gas combined-cycle plants or large-scale solar paired with battery storage, especially once financing costs and potential construction delays are factored in.
Developers can, and sometimes do, slow or pause construction in response to market shifts, supply chain problems, or design changes. Advanced reactors rely on specialized materials, components, and engineering expertise that are in limited supply. The federal documents verify regulatory progress but do not address whether vendors can deliver key systems on time or whether there are enough qualified welders, electricians, and nuclear engineers to staff multiple concurrent builds. If several advanced designs move from paper to construction at once, those practical constraints could become the binding bottleneck.
The Vogtle experience looms over every new nuclear project in the country. That expansion was also backed by federal loan guarantees, also championed as a new era for nuclear power, and also expected to come in on schedule. It did not. The Kemmerer and Hermes 2 permits are real and meaningful, but anyone tracking these projects should watch for the same warning signs: contractor disputes, component delivery delays, and cost estimates that creep upward between milestones.
What these permits actually tell us
As of spring 2026, the regulatory environment for advanced nuclear reactors in the United States is demonstrably more active than it has been in over a decade. Two construction permits have been granted for reactor technologies that did not exist in commercial form a generation ago. A third project is working its way through NRC review. Federal policy, from tax credits to licensing reform, is aligned behind new builds in a way it was not five years ago. And private-sector demand for clean, firm power is stronger than at any point since the original nuclear buildout of the 1960s and 1970s.
None of that erases the risks. Cost transparency is lacking. Construction timelines are undefined. Supply chains are untested at scale. The permits for Kemmerer and Hermes 2 break an eight-year dry spell in U.S. nuclear construction authorizations, and they prove that federal regulators can license new reactor technologies and that private developers are willing to invest in them. But these projects are still closer to the starting line than the finish. The real test is not whether they can win a permit. It is whether they can build a reactor on time, on budget, and prove that advanced nuclear power works not just on paper but on the grid.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
