Sensible Reform On the Way: Overview of the NRC Executive Order

On May 23, 2025, President Trump signed an executive order directing major reforms at the independent U.S. Nuclear Regulatory Commission, aiming to quadruple nuclear energy capacity by 2050. The order mandates faster, streamlined licensing and regulatory overhauls within 18 months, emphasizing reduced overregulation and a necessary NRC reorganization to handle the workload. While sweeping in ambition, the order avoids drastic specifics, leaving implementation details—and potential controversy—for future action and discussion.

On May 23, 2025, President Trump signed an executive order directing the reform of the U.S. Nuclear Regulatory Commission. This is highly unusual for an independent agency (executive orders are mentioned nowhere in the NRC’s own history report), and the idea that an executive order was coming had people in a tizzy for weeks. The actual order, while definitely worth serious consideration and something the NRC can’t hide from, is nowhere near the chainsaw that some were worried was coming (note: I wasn’t one of them). In this and subsequent articles, I’ll review the provisions in the order, as well as what they mean for U.S. nuclear energy and for Everstar.

What It Says

The order makes four key points.

1: The U.S. needs to be a global nuclear energy leader, so we are going to 4x nuclear capacity by 2050.

This is the general thread of the purpose and policy sections. While this position is advanced more substantively by the other nuclear EOs (national security, reactor testing, and industrial base), it is an important backdrop and “why” for the rest of the order.

2: Over-regulation can act as a drag on expansion of a beneficial technology.

This is the perennial argument regarding any regulator. Some think it’s over-regulating, some think it’s under-regulating. That’s one good reason that rules get put out for public comment, so that the NRC can hear all of those views and decide how to proceed. Though people disagree on exactly where the sweet spot is, there’s pretty universal agreement on a few points:

* regulation is needed to reassure the public that a hazardous technology is OK to have

* over-regulation is bad (if you're going to choke it, why allow nuclear at all)

* applying old, giant-reactor rules to little bitty reactors just feels wrong

3: The NRC needs to revise its rules within the next 18 months.

This is the meat of the EO, demanding a near-impossible schedule for a sweeping set of regulatory changes. Four articles will follow this one to illuminate important aspects of this massive rulemaking project:

* logistics of conducting such a “review and wholesale revision” and (importantly to us) how artificial intelligence can help

* easier parts of the rule related to deadlines, environmental changes already underway, restrictions on mid-stream changes to requirements, and a couple of policy measures that may not need rulemaking

* radiation protection standard changes, which are going to catch a lot of public attention but may not be very dramatic in the end

* streamlined licensing, which ties back to the key policy purpose of the four nuclear EOs and will link several of them together

4: The NRC must reorganize to provide appropriate focus on reactor licensing.

This is where people were hearing the chainsaws rev up. We can take a breath, though, and look at the massive amount of work coming the NRC’s way in both operating reactors (renewals, uprates, risk-informed initiatives) and new reactors (design approvals, license applications, infrastructure for broad-scale deployment). Of course they are going to have to reorganize, and probably hire, to get all this done.

The “reduction in force” piece is in there, and could be a little scary. Even with special hiring authorities given to the NRC in the ADVANCE Act, the fastest hiring is from within, by moving people around. The sensible thing to do after that is consider whether the holes left by those people need to be filled. If not—reduction in force. If so—smart staffing.

A future article will go deeper on this reorganization and the overall mission changes at the NRC, because it’s a big source of fear and hype.

What It Doesn’t Say

There is a lot that’s NOT in the order. Does that mean more are coming? Maybe. This is definitely plenty to chew on for now, though. A quick overview of what isn’t in the order:

1: Specific technical requirements.

It says to get license reviews done in 18 months and license renewal reviews done in 12 months. Almost everything else sets a general policy objective, but not how to accomplish it or any specific threshold. This should give some comfort to anyone who’s focused on changes to independent agency structures.

2: Direction on how to achieve the timelines.

This is more puzzling than the lack of technical requirements. There are a few things the EO could have done that would have made the rulemaking and licensing schedules more achievable:

Reductions in interagency reviews. OMB is invoked as a “working with” partner, but in reality OMB (again, very busy with many priorities) is often a schedule block at the end of a rulemaking or guidance process. The EO could have capped times on OMB reviews for Paperwork Reduction Act and Congressional Review Act, but does not. (It’s also not clear how this relates to the other EOs on deregulation and sunsetting.)

Strict engagement periods. The NRC often wrings its hands over whether to extend comment periods, have extra meetings, and otherwise go out of its way to engage a public that largely doesn’t care about what it’s doing. An EO could have given top cover to allow shorter (but still meaningful) engagement on rules and licenses.

Waiver of regulatory analysis. The NRC spends a LOT of time developing detailed cost-benefit analyses of every rule, following both EOs and internal policies, then arguing internally and externally about the details of the analyses. If the presumption of this EO is a reduction of regulatory burden, it could also have explicitly stated that detailed cost-benefit analyses are needed only when important to distinguish among options for action.

Timelines for Commission action. This is the age-old question at NRC all-hands meetings—when will the Commission act on X? The answer is always some version of “we have a lot on our plate and need to prioritize,” though the staff never gets to use this answer. The Commission will be involved in every rule change (unless it delegates) and every mandatory hearing (though it’s trying to speed that up). Though I know every one of the Commissioners and their staff personally, and I am confident they’re doing what they think is right, a little “heal thyself” may be in order.

I'm hopeful that the NRC can do some of this on its own authority.

3: Massive environmental changes.

It says to revise regulations to reflect the 2023 amendments to the National Environmental Policy Act, but this is already underway. As a matter of fact, legislative changes already bind the agency, so the NRC would need to exempt itself from any conflicting regulations until they are changed. It does point to the “Unleashing American Energy” EO, so that will need to be considered in the rulemaking.

However, it does NOT include sweeping language like in the reactor testing EO aimed at DOE: “use all available authorities to eliminate or expedite the Department’s environmental reviews for authorizations, permits, approvals, leases, and any other activity requested by an applicant or potential applicant. … [including] determining which Department functions are not subject to NEPA, creating categorical exclusions as appropriate for reactors within certain parameters (or relying on existing categorical exclusions), relying on supplemental analyses where reactors will be located on existing sites, or utilizing alternative procedures under NEPA.”

4: Anything international.

A lot of the talk about American dominance in both nuclear and AI (two related topics dear to Everstar’s heart) is centered on the premise the U.S. needs to get out there before other adversary countries do. The NRC already does a lot of international work to align its requirements with others’ and stand up regulatory bodies in the developing world, both of which facilitate global deployment. This was called out in Section 101 of the ADVANCE Act, so maybe nothing more was needed.

5: Non-reactor specifics.

The NRC does a lot beyond fission reactor licensing—fusion machines, radioactive materials users, fuel fabrication (the “front end” of the fuel cycle), fuel transportation and storage (the “back end”), and reactor decommissioning. These could be swept into a “wholesale review” of regulations and are partially within the scope of the NEIMA milestone schedules (certificates of compliance, which are about fuel, are specifically mentioned in the EO). However, it’s not likely they’ll be touched much, most practically because the rulemaking will already be unwieldy. People might wonder why fusion isn’t included, but that’s likely timeline-driven—shovels aren’t going into the ground on a commercial project anytime soon.

6: Certain other specifics.

As a policy matter, it says the U.S. should support the current fleet, indicates that fees are too high, and notes the NRC’s “leading reputation for nuclear safety.” However, there are no specific provisions that address these points, other than as potential outcomes of expedited schedules and modest reorganizations. It’s possible that future presidential or NRC actions could address things like fee reforms, operating-reactor initiatives, and workforce development.

What It Means

For the nuclear industry—a clear policy signal that safe expansion is supported, but without schedule demands or financial supports. (It’s the timelines in the other EOs, like reactors critical by next July, that will really have them racing along as long as DOE and DOD have enough budget.)

For the NRC—a reprieve from the gallows, but a warning that they need to hustle up licensing and have a good reason for any restriction they impose. They can’t just keep their heads down any more, because people will be watching.

For Everstar—opportunities for AI to assist with regulatory comparisons and commenting, preparation complete and well-organized applications that can be approved on short turnaround, optimization of application drafting and review schedules, and NRC review of applications and drafting of evaluations.

For the general public—if they’re even paying attention, not too much. The NRC still exists as a safety regulator and isn’t deleted. They’ll be distracted by rulemaking (more to come on that in a future article) and will be focusing their best people on licensing (I hope). As long as they can keep hiring, there’ll be plenty of smart people to do what the American public cares about.

What’s Coming Next

Five more articles are coming related to this EO, because the topics are so meaty. Watch this space for:

1. Burning Platform, Fiery Distraction - NRC Regulatory Overhaul Logistics

2. Piece of (Yellow)Cake - The Easier Parts of the NRC-Reform Rule

3. ALARA Alarm - Radiation Protection Standard Changes

4. Atomic Highway - Streamlined Reactor Licensing

5. Zero-Sum (or Less) Game - NRC Reorganization