The Regulatory and Siting Reality

Somewhere in Rockville, Maryland, a government building that most Americans have never heard of holds the keys to every nuclear reactor that will ever operate in the United States. The Nuclear Regulatory Commission occupies a position of extraordinary power in the energy landscape. It is the only federal agency that can say yes to a nuclear plant, and for decades, it was an agency whose institutional culture was far more comfortable saying no. In February 2026, the NRC announced a sweeping reorganisation, its chairman outlined mandatory 18 month licensing timelines under Executive Order 14300, and the agency simultaneously disclosed that it was managing over two dozen active rulemakings with a workforce that congressional testimony described as insufficient for the task. This is the regulatory reality that every SMR project must navigate, and understanding it is not optional for anyone deploying capital.

This module maps the complete regulatory terrain. We examine the NRC's new Part 53 framework, the pre-application process that precedes any formal licensing submission, the agency's capacity constraints, the staffing rules that were written for a world of single reactor plants, the emergency planning zone breakthroughs that change the siting calculus entirely, the physical security modernisation that could eliminate armed guard requirements, and the Illinois legislative revolution that has made the state the most attractive jurisdiction in America for nuclear data centre development. The audience is the infrastructure investor and the data centre operator who needs to understand which regulatory risks are real, which are solvable, and which have already been solved.

The Old Rules and Why They Broke

The United States built its nuclear regulatory framework in the 1970s around one type of reactor: the large light water reactor producing 1,000 megawatts or more. The regulations codified in 10 CFR Parts 50 and 52 were prescriptive, deterministic, and technology specific. They told applicants exactly what to build, exactly how to analyse it, and exactly what safety systems to install. This approach made sense when every reactor was a variation on the same pressurised water or boiling water theme. It makes no sense at all when the applicant is proposing a 15 megawatt microreactor using TRISO fuel in prismatic graphite blocks, or a 77 megawatt module sitting submerged in a below grade pool with no active cooling systems.

The regulatory mismatch created a practical problem that compounded over decades. Every novel design feature required an exemption. Every exemption required a safety evaluation. Every safety evaluation consumed NRC staff hours that applicants paid for at the full professional rate. The cumulative effect was a licensing process that could take four to six years for large light water reactors and an uncertain, potentially longer timeline for anything that did not fit the regulatory template. The Nuclear Energy Innovation and Modernisation Act of 2019 was Congress's acknowledgment that the framework needed fundamental reform. The result is Part 53.

What Is 10 CFR Part 53?

The NRC's risk informed, technology inclusive regulatory framework under 10 CFR Part 53 is scheduled for final rule publication on March 27, 2026, with an effective date of April 27, 2026. This is not an incremental update. It is a fundamentally different approach to nuclear licensing that replaces prescriptive design requirements with performance based safety objectives.

The implications for multi-module SMR deployment are significant. Part 53 uses the term "plant" rather than "reactor" to encompass configurations that Parts 50 and 52 were never designed to accommodate. It allows applicants to establish design specific staffing plans, control room configurations, and safety analysis methodologies tailored to the integrated operation of multiple modules from a single control facility. Rather than treating each module as a separate unit subject to redundant licensing reviews, Part 53 enables a unified licensing basis for the entire multi-module facility. It also explicitly addresses co-location with industrial facilities, concluding that the regulatory framework is sufficiently flexible to handle heat transport interconnections between nuclear plants and industrial operations, including data centres.

The most contentious element was the treatment of comprehensive risk metrics. The initial draft included Quantitative Health Objectives measured at efficiencies of one in a million risk per exposed individual. Stakeholders argued these were mathematically impossible to verify in real time operations. The Commission directed staff to remove them from the final rule, allowing applicants to propose their own comprehensive risk metrics tailored to their specific designs. This flexibility is a meaningful concession that benefits SMR developers whose designs may have risk profiles fundamentally different from large light water reactors.

Licensing Path	Framework	Best For	Timeline (est.)	Multi-Module Support
Part 50 (Two-Step)	Prescriptive, deterministic	Proven LWR designs	4 to 6 years	Exemptions required
Part 52 (One-Step COL)	Prescriptive, deterministic	Certified designs	3 to 5 years	Design certification specific
Part 53 (New)	Risk-informed, performance-based	Any reactor technology	Target 18 months	Built-in from inception
EO 14300 Mandate	Executive directive	All new reactors	18-month cap	Applies to all paths

Executive Order 14300, issued by President Trump in May 2025, overlays a hard constraint on the NRC: 18 month caps on licensing decisions for new reactors and 12 month caps for license renewals. Whether the agency can actually meet these timelines is one of the central questions facing the industry in 2026.

How Does NRC Pre-Application Engagement Work?

Before any formal licensing submission, the NRC has formalised a structured pre-application engagement process designed to reduce regulatory surprises and compress licensing timelines. The process is voluntary but strongly encouraged, and skipping it is a false economy that virtually guarantees delays during formal review.

The sequence begins with a Letter of Intent formally notifying the NRC of the applicant's plan to submit a licensing action for a specific site. This triggers assignment of a project tracking number and initiates the formal relationship. The applicant then develops a Regulatory Engagement Plan articulating its strategy for working with NRC staff on technical, regulatory, and administrative topics that may present challenges, involve novel approaches, or require Commission level policy decisions.

The core engagement consists of three mechanisms. White papers and technical reports allow applicants to request NRC feedback on novel approaches before formal submission. Topical reports request formal NRC staff review and approval of singular but critical technical issues, creating regulatory precedent that future applicants can reference rather than re-litigating the underlying technical basis. Readiness assessments evaluate the draft completeness of an application, identifying gaps before formal docketing triggers the cost and delay of Requests for Additional Information.

The ADVANCE Act, enacted in July 2024, made this process substantially cheaper. The reduced hourly rate for advanced nuclear reactor pre-applicants is now $148 per hour. That is a reduction of over 50 percent from the previous full cost professional rate of $318. An early site permit review involves approximately 29,000 NRC staff hours. At the reduced rate, that review now costs approximately $4.3 million. At the previous rate, the same review cost $9.2 million. Initial high level interactions, including introductory public meetings, are not billed at all. The reduced rate applies through September 30, 2030.

Pre-ADVANCE Act

Post-ADVANCE Act

NRC Hourly Rate ($)

The NRC's own testimony indicates that early mover advanced reactor developers have achieved construction permit review timelines of 17 to 18 months. Tennessee Valley Authority's Clinch River BWRX-300 project and Long Mott Energy's X-Energy project are both expected at those timelines. But this assumes comprehensive pre-application work has been completed prior to formal submission, which typically requires 18 to 36 months of engagement depending on design novelty and applicant preparation.

Can the NRC Actually Meet Its 18 Month Deadline?

The regulatory framework is modernising. The question is whether the agency has the people to execute it. The answer, based on Congressional testimony, NRC disclosures, and industry assessments current through February 2026, is deeply uncertain.

The NRC faces an unprecedented convergence of demands. It is simultaneously executing a major organisational restructuring announced on February 4, 2026, managing 25 advanced reactor vendors in active pre-application engagement, processing formal applications for NuScale, TerraPower, X-Energy, and others, and executing more than two dozen rulemakings mandated by Executive Order 14300.

“Restructuring while maintaining adequate staffing and improving NRC organizational culture is essential to continue enabling the path forward. Its ability to meet new licensing demands will depend on a workforce that is capable, accountable, motivated, and aligned at every level.”Judi Greenwald, President and CEO, Nuclear Innovation Alliance (January 2026 Congressional Testimony)

The reorganisation creates three core business lines: new reactors, operating reactors, and nuclear materials and waste. Each will integrate licensing and inspection functions to create a single point of accountability. The concept has theoretical merit. The timing is terrible. The announcement did not specify interim milestones, contingency plans, or metrics for assessing implementation progress. The agency said it would strive to implement the reorganisation by the end of September 2026.

The rulemaking workload adds another dimension to the capacity problem. A front loaded wave of proposed rules is scheduled between March and May 2026. A compressed cluster of final rules targets publication between September and November 2026. Critical rulemakings include licensing requirements for microreactors (proposed March 30, 2026, final September 16, 2026), the Part 53 risk informed framework (final March 27, 2026), and streamlined reviews of proven reactor designs (proposed April 2026). All of this during organisational restructuring.

A February 2026 NucNet report stated bluntly that job cuts at the DOE and NRC are hurting the goal of energy leadership, with the Nuclear Innovation Alliance president commenting that more staff are needed as industry activity grows. The NRC has not published a formal, quantitative staffing plan. Specific FTE additions, department level hiring timelines, and retention metrics remain undisclosed. This lack of transparency represents a critical information gap for investors evaluating the realistic timeline for advanced reactor licensing.

The honest assessment for project sponsors: the regulatory framework under Part 53 and EO 14300 is genuinely modernised and substantially faster than the historical baseline. But whether the agency can execute 18 month licensing reviews while restructuring, managing unprecedented rulemaking volume, and operating with a workforce that multiple sources describe as under resourced remains the single largest regulatory risk facing the SMR sector in 2026 and 2027.

The Staffing Problem: When Rules Were Written for One Reactor at a Time

The regulatory treatment of operator staffing for multi-module reactor facilities represents the most consequential gap between existing NRC regulations and the deployment requirements of the advanced reactor fleet. The problem is straightforward: the rules were written for a world where each reactor had its own control room and its own operators. SMR designs propose operating twelve modules from a single control room with three operators.

The foundational regulation, 10 CFR 50.54(m), prescribes minimum staffing by position and number of units. For a single unit with one control room: one Senior Operator and one Operator. For two units with one control room: one Senior Operator and two Operators. For three units, the regulation assumes at least two control rooms and mandates eight licensed operators. The regulation contains no provision for three or more units controlled from a single control room, which is precisely the configuration that NuScale and other SMR designs propose.

NuScale solved this through an exemption. In its Design Certification Application, NuScale requested that the NRC approve alternative staffing allowing up to 12 power modules to be operated from a single control room by a minimum shift crew of three licensed operators: two Senior Reactor Operators and one Reactor Operator. The NRC approved this in its January 2023 design certification, based on NuScale's passive safety characteristics, the simplicity of tripping a module and placing it in passive cooling mode, the fact that no operator intervention is required within 72 hours following defined accident scenarios, pilot operator training demonstrating adequate performance in multi-module events, and high fidelity simulator validation exercises.

The exemption pathway works. It is also expensive, slow, and design specific. Each advanced reactor vendor whose design has different passive safety characteristics, module independence, or automation levels must submit its own topical report, conduct human factors analyses, perform simulator validation, and justify its specific staffing level. Part 53 offers an alternative: rather than prescriptive staffing tables, it permits applicants to develop and justify design specific staffing plans through rigorous human factors analysis, provided they demonstrate the proposed staffing is adequate for all modes of operation and accident response.

Emergency Planning Zones: The Siting Revolution

If the staffing problem is a nuisance, the emergency planning zone question is transformational. For decades, every commercial nuclear reactor in the United States has been required to establish an Emergency Planning Zone with a 10 mile radius. This requirement, based on risk assessments of large light water reactors, dictates the area within which pre planned evacuation, shelter in place, and emergency notification strategies must be maintained. A 10 mile EPZ means sirens, dedicated alert systems, emergency drills, coordination with state and local governments, and fundamental restrictions on what can be built nearby. It is the single biggest constraint on nuclear siting near populated areas and industrial facilities.

NuScale has broken this constraint. On October 20, 2022, the NRC's Advisory Committee on Reactor Safeguards formally concurred that NuScale's methodology for determining EPZ size is acceptable for use by NuScale small modular reactor power plants. The technical finding: using NuScale's approved methodology, an EPZ limited to the site boundary of the power plant is achievable for a wide range of potential plant sites.

“NuScale's topical report provides a technically adequate method for assessing plume exposure pathway EPZ size for a NuScale SMR plant design. The methodology is generally consistent with the technical basis of the current 10-mile EPZ and there is reasonable assurance the methodology is adequate for sizing of the EPZ.”NRC Advisory Committee on Reactor Safeguards, Letter ML22287A155 (October 2022)

The operational significance is enormous. A site boundary EPZ means no 10 mile radius of emergency planning infrastructure. No community wide siren networks. No multi-county evacuation planning. No fundamental restriction on building data centres, industrial facilities, or commercial developments immediately adjacent to the plant boundary. A NuScale plant with a site boundary EPZ can accommodate siting of process heat off-takers, businesses, and housing in close proximity. It significantly reduces ownership costs for emergency preparedness. It makes industrial co-location not just possible but practical.

The technical basis rests on NuScale's demonstrable design safety. The passive cooling system requires no operator action, no AC or DC power, and no additional water to maintain safe conditions indefinitely following any defined accident scenario. The small core size and distribution of fuel across multiple independently contained modules reduces the potential source term. The below grade, pool immersed configuration provides inherent protection against external hazards. The NRC's methodology requires a full scope probabilistic risk assessment addressing both internal and external hazards across all operating modes. The result is that the radiological consequences at the site boundary, for the full spectrum of credible accident sequences, remain below the dose criteria that would require offsite protective actions.

EPZ Parameter	Large LWR (Current Fleet)	NuScale VOYGR	BWRX-300
Plume Exposure EPZ	10 mile radius	Site boundary	Under NRC review
Ingestion Pathway EPZ	50 mile radius	Proportionally reduced	TBD
Siren/Alert Network	Required (10 mile)	On-site only	TBD
Multi-County Coordination	Required	Not required	TBD
Industrial Co-Location	Severely restricted	Fully enabled	Expected to be enabled
NRC Methodology Status	Established (NUREG-0396)	Approved (Oct 2022)	Pre-application (Dec 2024)

The BWRX-300 has not yet received equivalent NRC approval. GE Hitachi is in active pre-application engagement on its source term methodology and EPZ framework. In a December 2024 pre-application meeting, GEH stated that the combination of the BWRX-300's safety strategy and inherent safety features makes the likelihood of significant nuclear fuel failure including core melt "not credible." No full power internal events within the proposed evaluation frequency result in core melt or appreciable fission product release into containment. The NRC staff provided detailed guidance on the proposed methodology and regulatory requirements. A formal source term topical report has not yet been submitted. Investors should expect the BWRX-300 to eventually achieve a reduced EPZ, but the formal NRC approval is likely 18 to 24 months behind NuScale's.

Physical Security: From Armed Fortress to Risk-Based Protection

The current nuclear security paradigm requires every operating reactor to maintain a minimum of 10 onsite armed responders at all times, along with redundant alarm stations, physical barrier systems, and comprehensive force on force exercises. These requirements were designed around large light water reactors containing enormous quantities of fissile material. Applying them unchanged to a 77 megawatt module sitting in a sealed steel containment submerged in a pool of water makes no engineering sense, and the NRC has acknowledged this.

On August 9, 2024, the NRC published a proposed rule (89 FR 65226) establishing alternative physical security requirements for advanced reactors. The final rule is expected in March 2026. The proposed rule introduces five specific alternatives available to qualified SMRs and non-light water reactors.

The eligibility criterion is consequence based rather than prescriptive. An applicant must perform a technical analysis demonstrating that the offsite radiological dose consequences of a postulated, unmitigated design basis threat attack, considering only facility design and siting, are at or below 25 rem total effective dose equivalent at the exclusion area boundary. If the reactor's inherent design means that even an unmitigated security attack cannot produce offsite consequences exceeding that threshold, the full security apparatus designed for large reactors becomes unnecessary.

The proposed rule also redefines the security objective for advanced reactors. For large reactors, the physical protection programme must prevent significant core damage and spent fuel sabotage. For advanced reactors, the objective shifts to preventing a significant release of radionuclides from any source. This is a technology inclusive formulation that does not assume the failure modes of any particular reactor type.

Why Is Illinois the Most Nuclear Friendly State in America?

While the federal regulatory framework determines whether a reactor can be licensed, state law determines whether it can be built. Until January 2026, Illinois prohibited construction of any nuclear reactor with nameplate capacity exceeding 300 megawatts. That prohibition is now gone. Senate Bill 25, the Clean and Reliable Grid Affordability Act, signed by Governor Pritzker on January 8, 2026, represents the most comprehensive overhaul of Illinois energy policy since the 2021 Climate and Equitable Jobs Act.

The legislative history matters for understanding the state's trajectory. In August 2023, Governor Pritzker vetoed an earlier bill lifting the large reactor moratorium, citing concerns about "costly" large reactors causing "exorbitant ratepayer funded bailouts." By January 2026, his position had reversed completely. At the signing ceremony in Joliet, he stated: "In Illinois, we are pursuing every available option to produce affordable, efficient, clean, and abundant energy. We are leaving no stone unturned in the work to produce more electricity, lower prices for our people, and secure our long-term energy future."

The reversal reflects the same fundamental calculus driving the hyperscaler nuclear commitments documented in Module 1. Growing electricity demand from AI infrastructure requires baseload generation that renewables alone cannot provide. Illinois, which hosts 11 operating reactors at six sites and produces more nuclear generated electricity than any other state, possesses institutional knowledge, workforce expertise, and grid integration infrastructure that no other state can match.

SB 25 does more than lift a moratorium. It grants the Illinois Commerce Commission unprecedented authority over long term energy resource planning through new integrated resource plan mechanisms covering 5, 10, 15, and 20 year horizons. Utilities must project energy demand across these planning periods and include detailed modelling on emissions, affordability, equity, and grid reliability. The ICC can approve, modify, or reject resource plans. This means the ICC could effectively mandate inclusion of nuclear generation in utility long term plans, creating a demand signal that de-risks developer investment.

The Illinois Power Agency projects that the critical measures in SB 25 will save Illinois energy customers $13.4 billion over two decades. The Illinois Manufacturers' Association disputes this. Its counter-estimate covers a large auto manufacturer using 10,100 kilowatts. That manufacturer would see first year monthly rate increases of $11,361. By 2045, the association projects the monthly increase rises to $87,276. The truth likely falls between these projections, but the direction of state policy is unmistakable: Illinois is positioning itself as the jurisdiction of choice for nuclear energy development, and it has the existing nuclear infrastructure to make that positioning credible.

The Illinois Nuclear Ecosystem

The state's attractiveness extends beyond legislative reform. Illinois is actively building an advanced reactor ecosystem that combines existing fleet operations, academic research, and advanced manufacturing.

Constellation Energy operates the state's existing fleet. In December 2025, the NRC granted license renewals for Clinton (operating through 2047) and Dresden (through 2049 and 2051). Constellation invested over $370 million in relicensing capital improvements. Meta signed a 20 year power purchase agreement with Constellation to ensure Clinton's continued operation, directly linking hyperscaler demand to Illinois nuclear assets.

NANO Nuclear Energy has raised over $600 million since its May 2024 IPO. It is establishing a manufacturing and research facility in Oak Brook, Illinois. The facility is supported by $6.8 million in state incentives through the REV Illinois programme. The company acquired a 23,537 square foot facility for the project. That facility includes a dedicated 7,400 square foot non-nuclear demonstration area. The University of Illinois Urbana-Champaign plans to submit a construction permit application for the KRONOS Micro Modular Reactor to the NRC in the first quarter of 2026. This 45 megawatt thermal, 15 megawatt electrical high temperature gas cooled reactor using TRISO fuel will partially re-power the university's existing Abbott power station, providing a zero carbon demonstration of district heat and power to campus buildings.

The KRONOS project is pursuing Part 50 two step licensing. Site characterisation and geological drilling work was completed in November 2025 with AECOM. The university has received NRC safety evaluations on multiple topical reports. If the construction permit application proceeds on schedule, Illinois will host the first campus scale microreactor demonstration in the United States, creating a proof point for the industrial co-location model that data centre operators need to see.

How Long Does It Take to License a Nuclear Data Centre?

For the infrastructure investor evaluating a nuclear data centre project in 2026, the regulatory timeline maps to a specific sequence of decision points and capital commitments. The process is long but it is knowable, and the recent reforms have compressed it meaningfully.

Phase	Duration	Key Activities	Estimated Cost
Pre-Application Engagement	18 to 36 months	Letter of Intent, white papers, topical reports, readiness assessment	$4 to $10M (NRC fees + consultants)
Construction Permit / COL Application	17 to 18 months (NRC review)	Formal application submission, NRC staff review, ACRS review	$15 to $30M (application + review fees)
Public Hearing	3 to 12 months	Mandatory hearing, contested if intervenors	Variable (legal costs)
Construction	3 to 5 years (SMR)	Factory fabrication, site preparation, assembly	See Module 4 economics
Operating License / Startup	6 to 18 months	Fuel loading, testing, commercial operation	Included in construction
Total: Site Selection to Power	5 to 8 years	Full lifecycle from decision to electrons	Design dependent

The timeline breaks down into three phases. Pre-application engagement runs 18 to 36 months and costs $4 to $10 million in NRC fees and consultant expenses at the reduced ADVANCE Act rate. The construction permit or combined license review targets 17 to 18 months under EO 14300's mandate, at a cost of $15 to $30 million including application preparation and NRC review fees. Construction runs 3 to 5 years for SMR designs. The total from site selection decision to commercial power delivery is 5 to 8 years.

The critical insight is that the nuclear licensing timeline, while longer than gas turbine deployment, is comparable to or shorter than grid interconnection timelines in the markets where data centre demand is most acute. In Columbus, Ohio, grid interconnection takes 84 months. In Northern Virginia, Dominion Energy has 70,000 megawatts in queue with no connection dates for 45,000 megawatts of that total. A nuclear licensing process that delivers power in 5 to 8 years is not a disadvantage relative to a grid connection that takes 7 years. It is an alternative path that arrives at approximately the same time with fundamentally better economics, zero carbon emissions, and 20 year price certainty.

The regulatory landscape has changed more in the past 18 months than in the previous two decades. Part 53, Executive Order 14300, the ADVANCE Act fee reductions, NuScale's site boundary EPZ, the proposed physical security modernisation, and Illinois's comprehensive legislative reform have collectively transformed the regulatory environment from an obstacle to a navigable pathway. The remaining risk is execution: whether the NRC can deliver on the timelines its own leadership has mandated, with the workforce and organisational capacity it currently possesses. Module 6 takes every data point from this series, every economic model from Module 4, and every regulatory timeline from this module, and designs three fully specified nuclear data centre campuses that show what the end state actually looks like.