The answer is more complicated than it used to be, as multiple federal agencies are responsible for new nuclear reactors.

If you have been trying to follow recent headlines about nuclear reactor development in the United States and feel confused, you are not alone, and your confusion is well earned. One news story announces the granting of hundreds of millions of dollars in new federal funding under programs with names like ARDP (Advanced Reactor Demonstration Program) or RPP (Reactor Pilot Program). Another reports that a reactor has been “authorized” at Idaho National Laboratory or Oak Ridge, Tennessee. A third describes a construction permit moving forward at the Nuclear Regulatory Commission. Still another mentions the Department of Defense sponsoring a microreactor for use at a military base. 

Each story invokes an alphabet soup of acronyms — DOE, NRC, DoD, LPO, PTC, PPA, GAIN, FOAK — and each sounds, in its own way, like progress. But it can be difficult, even for people within the nuclear industry, to determine exactly what is happening. Has a reactor been approved to operate? Has it merely been funded? Is it licensed? And which federal agency, exactly, is exercising legal authority?

Until recently, the answer to those questions was much simpler. For most of the civilian nuclear era, approval of new reactors in the United States revolved around a single institution: the Nuclear Regulatory Commission. If a reactor was going to be built and operated, especially to generate electricity for the grid, it required a Nuclear Regulatory Commission license issued under the Atomic Energy Act. That basic framework still exists. But it no longer tells the whole story.

Today, authority over reactor deployment in the United States is no longer concentrated in one agency. Instead, it is distributed across three: the Department of Energy (DOE), the Department of Defense (DOD), and the Nuclear Regulatory Commission (NRC). Each acts under different statutory mandates, uses different regulatory tools, and supports or approves different kinds of nuclear activity. The resulting system is not incoherent, but it is complex and layered. And for those trying to assess the true state of American nuclear progress, it is often confusing.

From One Nuclear Energy Gatekeeper to Three

For decades, the structure of nuclear approval aligned neatly with the dominant use case for reactors: large, light-water plants built by utilities to sell electricity into regulated or wholesale markets. In that world, the NRC’s licensing authority under the Atomic Energy Act, implemented through Parts 50 and 52 of Title 10 of the Code of Federal Regulations (C. F. R.), was the central gateway. Other agencies mattered, but they played secondary roles.

That world is in the process of changing. Modern advanced reactors are smaller, more diverse in design, and often intended for purposes other than generating electricity. Some are built primarily to demonstrate technology. Others are aimed at industrial heat, remote or facility-specific power, or defense resilience. At the same time, federal policymakers have placed renewed emphasis on speed, learning-by-doing, and strategic competition. As a result, the United States has increasingly come to rely on multiple federal pathways to move nuclear projects forward. The NRC remains central, but it is no longer alone.

The DOE and the Return of Federal-Site Nuclear Authorization

The most significant shift has occurred at the Department of Energy. DOE does not license reactors in the NRC sense. It cannot grant a commercial operating license under the Atomic Energy Act. Yet DOE has always possessed something that, in certain contexts, is equally consequential — statutory authority over federal nuclear facilities and the authority to fund nuclear projects and approve nuclear reactors located on them.

When reactors are sited at national laboratories or other DOE-controlled locations, they are typically authorized under DOE’s own nuclear safety framework, most notably 10 C.F.R. Part 830. That regime requires a documented safety analysis, hazard categorization, and an approved safety basis and is subject to DOE oversight and enforcement. This is a binding regulatory authorization system, but it is distinct from NRC licensing, and it reflects the DOE’s separate role as both site owner and safety authority. These projects are also subject to review under the National Environmental Policy Act (NEPA), which imposes additional procedural and disclosure obligations.

This is not a regulatory loophole. It is a parallel legal framework that has existed for decades but that is now being used more actively as a reactor deployment mechanism.

The DOE has recently coupled this site-based authority with an increasingly robust portfolio of funding and risk-reduction programs. The Advanced Reactor Demonstration Program (ARDP) was designed explicitly to support first-of-a-kind and early follow-on reactor projects through cost-shared federal funding. The Reactor Pilot Program (RPP), created by Congress, similarly authorizes DOE to support the siting and construction of demonstration reactors. The Gateway for Accelerated Innovation in Nuclear (GAIN) program facilitates access to DOE expertise and infrastructure. The Loan Programs Office (LPO), tax credits, and power-purchase-agreement-style arrangements further shape the financial environment in which projects move forward.

Recent DOE funding selections involving utilities like TVA and companies like Holtec are best understood in this context. These decisions do not, by themselves, authorize the construction or operation of a commercial reactor. What they do is reduce first-mover risk, stabilize project financing, and signal federal commitment to particular deployment pathways.

The DOE’s growing focus on HALEU fuel infrastructure fits the same pattern. Fuel availability has emerged as one of the most significant constraints on advanced reactor deployment. By supporting enrichment, transportation, and qualification efforts, the DOE is addressing a systemic bottleneck that lies largely outside the NRC’s licensing purview but directly affects whether licensed reactors can ever operate.

The cumulative effect is that the DOE has reasserted itself as an enabling authority for nuclear reactors, particularly at federal sites and in the demonstration phase. To an outside observer, this can look like licensing. Legally, it is authorization of a different kind, rooted in DOE’s role as facility operator, safety regulator, and funder.

DoD and the Rise of Mission-Driven Reactors

Overlaying this DOE resurgence is a second, even more recent development: the growing involvement of the Department of Defense.

The DOD’s interest in nuclear energy is not driven by civilian electricity markets. It is driven by mission requirements — energy resilience, remote operations, and the assurance of power in contested or austere environments. Under its own statutory authorities, the DOD may procure energy systems necessary to support national defense missions. When nuclear reactors fit that need, the DOD may sponsor their development. This role is not new. Indeed, the Department of Defense and its predecessor institutions have been deeply involved in the development, approval, and operation of nuclear reactors since the earliest days of the nuclear age. 

The US Navy’s nuclear propulsion program, which predates the creation of the NRC and operates outside the civilian licensing framework, has long maintained its own rigorous safety, training, and oversight systems for nuclear-powered submarines and aircraft carriers. Today’s defense-sponsored microreactor initiatives are best understood not as a departure from historical practice, but as a modern extension of a long-standing model in which nuclear reactors are developed and operated under defense authority for mission-critical purposes.

Project Pele, a mobile microreactor initiative sponsored by the Strategic Capabilities Office, illustrates this model. Pele is not intended to sell power to the grid. It is designed to provide resilient energy for defense applications. The demonstration is occurring at Idaho National Laboratory, with the DOE providing the site and exercising nuclear safety authority. The NRC has been consulted, but it is not issuing a traditional commercial license for the project.

This creates a third category of nuclear activity in the United States: reactors that are neither NRC-licensed commercial plants nor purely civilian DOE research projects but defense-sponsored systems operating under federal control. 

The DOD is not becoming a nuclear regulator in the traditional sense. But its growing role underscores a broader shift in how reactors are increasingly viewed — not only as power plants, but as strategic infrastructure.

The NRC Remains Central — Especially for Commercial Power

Amid this proliferation of federal activity, it is essential not to misread the moment. The NRC has not been sidelined. For reactors that are intended to operate outside federal facilities, and especially for those meant to provide commercial electricity, NRC licensing remains the decisive legal gateway.

That authority rests squarely on the Atomic Energy Act and is implemented primarily through 10 C.F.R. Parts 50 and 52. Construction permits, operating licenses, combined licenses, early site permits, and design certifications all flow through this framework. These approvals carry legal durability, public legitimacy, and investor confidence that no other federal pathway can replicate.

The NRC is also far from inactive. It is currently processing a new generation of advanced reactor applications, including utility-led SMR projects, industrial-host reactors, test reactors such as Kairos Power’s Hermes facility, and design approvals like NuScale’s uprated upgraded module. The NRC’s authority over operating reactor decisions, including plant restarts, further underscores its continued centrality.

In short, while the DOE and DoD increasingly act upstream — enabling, funding, and authorizing demonstrations — the NRC remains the only institution that ultimately determines whether most reactors can be deployed commercially and at scale.

Why Nuclear Authority Has Been De-Concentrated — and Why This Is Contested

The shift from a single-agency model to a distributed system did not happen by accident. It reflects deliberate policy choices made over multiple administrations, driven largely by concerns about institutional fit and effectiveness. But it is also a shift that has generated serious and reasonable debate.

For much of the post-Cold War period, the NRC functioned as the practical chokepoint for nearly all reactor activity. That made sense when reactors were large, standardized, and built almost exclusively for grid electricity. To some observers, however, it has made less sense as new use cases have emerged — i.e., microreactors, demonstrations, defense applications, and first-of-a-kind systems intended primarily for learning.

Policymakers have increasingly questioned whether a single civilian regulator, optimized for commercial licensing, was the right institution to manage every category of nuclear activity. These concerns were often framed in terms of efficiency, and not in the sense of cutting corners. Rather, they reflected questions of alignment, i.e., whether different missions should be handled by different institutions already empowered by law to pursue them.

At the same time, critics of this shift have raised important concerns. Some argue that moving reactor projects outside the NRC’s traditional licensing framework risks eroding safety culture, fragmenting oversight, or reducing public confidence. Others worry that the DOE and the DOD, as mission-oriented agencies, may face institutional pressures that differ from those of an independent safety regulator. These critiques are not fringe views; they reflect longstanding debates about nuclear governance in the United States.

The policy response of multiple administrations has not been to strip authority from the NRC but to use the existing DOE and DOD authorities more assertively in parallel. By choosing federal sites, federal ownership, or defense missions, administrations have advanced nuclear projects under legal regimes arguably better suited to achieving those objectives without amending the Atomic Energy Act or eliminating NRC oversight where it remains legally required.

At the same time, successive administrations have continued to press the NRC to modernize and adapt, emphasizing greater predictability, transparency, and timeliness in licensing while reaffirming the Commission’s core safety mandate. Recent executive orders have reinforced this pressure by directing the NRC to review and, where appropriate, revise its licensing processes for advanced reactors with an explicit focus on reducing unnecessary delay without compromising safety. 

In response, the NRC has undertaken its own regulatory reform efforts, most notably through the development of Part 53, a new, technology-inclusive licensing framework intended to provide an alternative pathway for advanced reactors alongside the existing Part 50 and Part 52 regimes. Whether Part 53 will ultimately succeed in delivering more efficient licensing while maintaining public confidence remains an open question, and it is one that continues to generate debate among regulators, industry representatives, and public-interest groups. 

The intended result, however, is not a weakened NRC, but a more clearly bounded and adaptable one: i.e., still central to commercial nuclear deployment, but no longer the sole institutional gateway for every category of reactor activity in the United States.

A Trifurcated System — and How to Read It

What can appear as regulatory confusion is better understood as functional specialization. The United States now operates with three overlapping but distinct nuclear pathways. The DOE authorizes and enables reactors at federal sites and through federal funding. The DOD sponsors reactors to meet national security missions. The NRC licenses reactors for civilian commercial operation.

Headlines often blur these distinctions. But they matter. Funding is not licensing. Authorization at a national laboratory is not permission to sell power to the grid. And a defense demonstration is not a commercial approval.

For those trying to understand the true state of American nuclear progress, recognizing which agency is acting, and under which legal authority, is essential. The story is not that the United States has abandoned its traditional regulatory framework. It is that it has layered new federal tools on top of it.

That distinction is the key to understanding who authorizes new nuclear reactors in the United States today.

About the Author: Daniel Joyner

Daniel Joyner is the Elton B. Stephens Professor of Law at the University of Alabama School of Law and the founder and principal at Prometheus Nuclear LLC. Prior to joining the Alabama Law faculty in 2007, Dr. Joyner taught for four years on the faculty of the University of Warwick School of Law. He is the author of International Law and the Proliferation of Weapons of Mass Destruction (Oxford University Press, 2009); Interpreting the Nuclear Nonproliferation Treaty (Oxford University Press, 2011); and Iran’s Nuclear Program and International Law (Oxford University Press, 2016).

Image: Shutterstock/kittirat roekburi

The post Who Authorizes New Nuclear Reactors in the United States? appeared first on The National Interest.