We all know that second chances are rare. Those that make theirs count study the moment while everyone else is celebrating it, and they move with intent before the rules are obvious to anyone else.

Nuclear is living through its second chance right now. Money and political will have both returned after a decade away, and public opinion has moved with them. The momentum is loud. It is producing real projects at real speed. Alongside the announcements, a steadier instinct showed up this week. The decisions that surfaced were the deliberate kind, the moves an industry makes when it intends to still be standing years from now.

Every one of this week's moves took a risk off the table and left the largest commitment for later. That is a particular kind of discipline. An industry retires risk ahead of the payoff only when it fully expects to collect. Five separate parts of the sector made that same move within days. The most literal version is in Virginia, where a utility just paid to keep its options open while it decides which reactor to build. Start there.

Dominion Energy Virginia filed in July to renew Early Site Permit ESP-003 for the North Anna site in Louisa County, the permit that qualifies the ground for new reactors. It dates to November 2007 and expires in November 2027, and Dominion wants twenty more years on it. Alongside the renewal, the utility asked the NRC to waive the subsumption rule in 10 CFR 52.26, so the full site permit stays intact and in play. It also formally deferred the combined license it already holds, NPF-103, granted in 2017 for an Economic Simplified Boiling Water Reactor (ESBWR). No construction has started. Dominion's stated reason for the waiver is that it may license a design other than the ESBWR, and it wants the whole site open for whatever it picks. Here is the move under the paperwork. A combined license names a reactor; a site permit qualifies the land. Dominion is renewing the slow, hard-won thing and keeping the reactor choice open while it decides, because the qualified site is the asset, and the machine can come later. [1]

The next move came in Texas, where the risk being retired was the design itself. The Department of Energy approved a Nuclear Safety Design Agreement for Natura Resources, clearing the first gate of the pathway that will let its molten salt reactor be built and run. The machine is small, a 1 MWt research reactor bound for Abilene Christian University. It is one of eleven projects in the DOE pilot program, the pathway that authorizes advanced reactors at sites outside the national labs. The agreement fixes the safety analysis approach and names the key safety decisions before any construction. That order matters more than usual. In a molten salt reactor the fuel is dissolved in the coolant, so the safety case cannot borrow from the solid-fuel light-water playbook that every operating US plant runs on. Approving the safety framework first is where the genuinely new engineering gets settled. Natura logged 1,000 hours on a flowing-salt test loop in the spring, so the paper is tracking real hardware. [2]

Money moved next, and it moved in bigger blocks. Mergers and acquisitions across the global nuclear industry reached $7 billion in the first half of the year, double the same period a year earlier, according to research by the law firm White & Case. The deal count barely changed, from 40 to 44. Value doubling on almost the same number of transactions means the average deal grew, and size is the story. The firm credits demand for data-center baseload, governments treating nuclear as energy security, and new funding models that share risk between public and private balance sheets. Ximena Vasquez-Maignan, the firm's nuclear practice lead, put it plainly, that financing is now sophisticated enough to allocate risk and improve what she called bankability. Buyers position before they build. Doubling the money in play while the number of deals holds steady is what it looks like when serious capital decides the sector is worth owning. [3]

The most striking result of the week happened at a reactor almost no one outside research has heard of. For the first time, a team adjusted a reactor's power remotely and autonomously in real time, and the reactor's own safety systems held final authority the whole way through. The reactor was PUR1, a low-power research unit at Purdue. The work spanned three live sites. High-performance computing sat at Idaho National Laboratory, the reactor ran in Indiana, and the control software ran in a Microsoft Azure cloud environment in Virginia. From Idaho Falls, a reinforcement-learning model calculated and sent the movements of an auxiliary control rod through INL's data platform, trimming power fluctuations with no one touching a rod on site. The risk this retires is operational. Automation is how a growing fleet holds down human error and running cost, and it earns a license only when the safety layer keeps the final say. This run proved the assist can happen while an independent safety system keeps the power to say no. The industry settled that question on a research reactor, well before it has to settle it on a fleet. [4]

The last move secured fuel. The Department of Energy made conditional commitments to supply high-assay low-enriched uranium to NASA and to Radiant Industries, the third such round from its allocation program. NASA's share goes to the SR-1 Freedom mission to Mars; Radiant's goes to a microreactor headed for Buckley Space Force Base in Colorado. HALEU has no domestic commercial supplier yet, and many advanced reactors need it to run at all. So DOE is drawing on government stock, including material from the National Nuclear Security Administration, to bridge the gap. A finished advanced reactor with no HALEU is a machine that cannot start, so a fuel allocation is the step that turns a licensed design into something deployable. NASA is a first-time recipient, which says where the priority now points. [5]

The momentum is real, and we have explored it across many editions this year. Groundbreakings, new projects, and fresh announcements arrive almost every week, and the acceleration is showing up as steel in the ground and signatures on contracts. What matters sits underneath the noise. Even in an era built for big announcements, the industry is still taking deliberate, well-considered steps, the kind its own history taught it to value. This week put five of them on the record. Each one protects a hard-won position and sets up the next decision to be made well. That care in the middle of a boom is a form of confidence. You handle a moment this carefully only when you fully intend to be standing at the end to collect on it.

These decisions share a second trait, one no single block could show. Each lowers the cost of a choice not yet made. A renewed and unbound site is ground the next build can start from, whatever design finally wins. Approve one molten salt safety case, and the next reactor of its kind arrives with a reference already written and part of the record. Fuel allocated now seeds the supply later reactors draw on. The work compounds. A good moment becomes a durable one this way, each careful call making the next one cheaper to get right.

Second chances reward the ones who move fast and still move well. Nuclear is doing both, breaking ground where it can and laying the groundwork with care beneath it. That is how a moment becomes a foundation.

This week the industry got ready for a moment it can feel coming. So here is my question for you: what are you keeping ready for the moment you see coming in your own work, before the rest of the field does?

More next week.

Dive deeper

  1. Dominion Files to Renew North Anna Early Site Permit and Defer Its Reactor Choice Dominion Energy Virginia filed in July to renew Early Site Permit ESP-003 for the North Anna site in Louisa County, seeking twenty more years on the permit that qualifies the ground for new reactors. The permit dates to 2007 and expires in November 2027. Alongside the renewal, the utility asked the NRC for an exemption from the subsumption rule in 10 CFR 52.26 so the full site permit stays intact, and it formally deferred the combined license it holds, NPF-103, granted in 2017 for an ESBWR. No construction has started. Dominion states it may license a design other than the ESBWR and wants the whole site open for whatever it selects.
  2. Molten Salt Reactor Gets US Safety Design Approval The US Department of Energy approved a Nuclear Safety Design Agreement for Natura Resources, clearing the first gate of the pathway that will let its molten salt reactor be built and operated. The reactor is a 1 MWt research unit bound for Abilene Christian University, one of eleven projects in the DOE Reactor Pilot Program that authorizes advanced reactors at sites outside the national labs. The agreement fixes the safety analysis approach and names the key safety decisions before construction. In a molten salt reactor the fuel is dissolved in the coolant, so the safety case cannot borrow from the solid-fuel light-water playbook. Natura logged 1,000 hours on a flowing-salt test loop in the spring.
  3. Global Nuclear M&A Deals Double to $7 Billion Mergers and acquisitions across the global nuclear industry reached $7 billion in the first half of the year, double the same period a year earlier, according to research by the law firm White & Case with Mergermarket. The deal count barely moved, from 40 to 44, so the average transaction grew and size is the story. The firm credits demand for data-center baseload, governments treating nuclear as energy security, and new funding models that share risk between public and private balance sheets. Ximena Vasquez-Maignan, the firm's nuclear practice lead, said financing is now sophisticated enough to allocate risk and improve bankability.
  4. Researchers Achieve Remote, Autonomous Power Control of a Research Reactor in Real Time A team adjusted a reactor's power remotely and autonomously in real time for the first time, with the reactor's own safety systems holding final authority throughout. The reactor was PUR1, a low-power research unit at Purdue. The work spanned three live sites. High-performance computing ran at Idaho National Laboratory, the reactor ran in Indiana, and the control software ran in a Microsoft Azure environment in Virginia. A reinforcement-learning model calculated and sent the movements of an auxiliary control rod through INL's data platform, trimming power fluctuations with no one touching a rod on site. The demonstration ran on a research reactor well before the question has to be settled on a fleet.
  5. Energy Department to Distribute Third Round of HALEU to NASA and Radiant The US Department of Energy made conditional commitments to supply high-assay low-enriched uranium to NASA and to Radiant Industries, the third round from its allocation program. NASA's share goes to the SR-1 Freedom mission to Mars, and Radiant's goes to a microreactor headed for Buckley Space Force Base in Colorado. HALEU has no domestic commercial supplier yet, and many advanced reactors need it to run, so DOE is drawing on government stock, including material from the National Nuclear Security Administration, to bridge the gap. NASA is a first-time recipient.

Process note: This brief is created using an AI-assisted workflow and reviewed before publication. Learn more about Finding Critical Path and how each edition is built at About — FindingCriticalPath.com.