I’m able to believe that the newest small reactors are much safer than previous designs.
What I continue to do is calibrate expectations on what these reactors will do, how much they cost, and when they will be available.
Given the substantial upfront capital investment required for deployment of new nuclear generation (relative to other types of generation), support will be required in the form of cost overrun protection, which currently does not exist, or other cost mitigation measures. Units 3 and 4 at Plant Vogtle were the first advanced reactor LLWR projects in the U.S. with two AP1000® units constructed and placed into commercial operation.
The Vogtle project experienced significant delays and cost overrun, with the total cost more than double the original projected cost. Though some issues were unique to that project, the first and second movers for the next advanced reactor projects will be assuming construction risks and therefore will need some form of insurance to protect customers.
It is essential that the current tax credits that incentivize new nuclear generation remain in place, as they provide critical financial support for these investments and decrease investment risk, which will lead to lower overall costs for customers. In addition, the current tax credits that incentivize the efficientoperation of the existing nuclear fleet are also essential, as they help keep energy prices low, with every dollar going back to customers. Further, the federal loan guarantee program, federal grant opportunities, and other mechanisms like public-private partnerships could provide crucial support.
Given the risks and long-lead time associated with constructing new nuclear, cost overrun protection would further decrease investment risk in a manner that would benefit customers.
In sum, without continued federal support, the financial risk of nuclear construction could deter development, slowing, delaying, or eliminating progress despite the substantial potential benefits of new nuclear generation for customers. Importantly, because Duke Energy is a regulated utility, federal support directly benefits customers in the form of lower rates and lower overall financing costs.
The top federal official in charge of giving loans to nuclear plants had some harsh words this week for nuclear industry officials who want the government to invest more taxpayer funds into building reactors.
“At a certain point, it’s just greedy,” said Julie Kozeracki, the acting chief investment officer for President Donald Trump’s Office of Energy Dominance Financing.
Kozeracki was speaking in front of hundreds of nuclear industry CEOs and financiers on a panel at the Nuclear Energy Institute’s finance conference in New York on Tuesday. She told them the industry needs to stop asking for new funding streams from the government and start building reactors. “The tools are there, and it’s time to deliver,” she said.
Trump has called for the U.S. to start construction on 10 large nuclear reactors by 2030, to produce enough electricity to meet growing American energy demands, including the power needed by artificial intelligence data centers. That goal is likely to cost well over $100 billion, almost all of it paid by private companies.
Meanwhile, the plan also envisions at least 4000 MW of solar by 2034.
Energy Secretary and Fracking Baron Chris Wright keeps saying that nuclear plants will be “going critical” this summer, as if this means that some nuclear renaissance is fully underway, but there is less here than meets the eye.
I asked two AIs for a fact check.
Google Gemini:
Energy Secretary Chris Wright’s announcement refers to microreactors and advanced small modular reactors (SMRs) that are hitting critical milestones ahead of the July 4th deadline mandated by President Trump’s Executive Order 14301.
The exact reactors, their sizes, and their power-generation timelines are detailed below.
- What Reactors Are They?
The Energy Department is tracking up to four distinct, privately developed “non-light-water” test reactors hitting criticality under the DOE Reactor Pilot Program:
- Antares Nuclear “Mark-0”: This reactor officially became the first to achieve “zero-power” criticality at the Idaho National Laboratory (INL) on June 4, 2026. It uses a sodium heat-pipe-cooled system and high-assay low-enriched uranium (HALEU) TRISO fuel.
- Aalo Atomics “Aalo-X”: A sodium-cooled test unit currently targeting first criticality.
- Radiant Nuclear “Kaleidos”: A helium-cooled microreactor preparing for testing inside INL’s renovated dome facility.
- Valar Atomics: Another selected pilot reactor design that received preliminary safety design approval to test.
- How Large Are They?
These are not massive traditional power plants; they are ultra-compact microreactors and miniature SMRs designed to fit on trucks and inside shipping containers.
- The Test Units: The immediate prototypes (like the Mark-0) are “zero-power” or thermal-only units. They generate heat for physical validation rather than electricity.
- The Commercial Scale: When these models scale up to commercial production, they will remain small. For example, Radiant’s Kaleidos targets 1.2 Megawatts (MW) of output, while Aalo’s subsequent commercial design (the Aalo-1) targets 50 Megawatts-electric (MWe).
- When Will They Generate Significant Power?
While reaching criticality means the nuclear chain reaction is self-sustaining, none of the reactors turning on this summer will generate immediate grid power.
- Late 2027: Antares Nuclear expects its subsequent iterations to begin producing initial electricity by late 2027.
- 2028: Commercial site deployments and field operations for microreactors like Antares and Radiant are slated to begin.
- 2029–2030: Commercial grid-scale versions, such as Aalo Atomics’ 50 MWe
ChatGPT:
Not sure about the others – I think they’re mostly niche – but Aalo Atomics is aiming for power below $30/MWh, marginally better than gas and established nuclear. Their test reactor is the same physical size as their proposed commercial unit, just run at 1 MW to check things, not the target 10 MW (to be batched in fives per turbine.) They use standard low-enriched fuel, but at low pressure, so don’t need heavy forged pressure vessels or expensive HALEU uranium/Triso. Also don’t seem to be sucking up to Trump as much as the Valar guy, which is a plus (CTO is a Bengali, CEO’s Canadian, so nativism is out.)
“Aiming” at a price point is all very well and good, but we’ve seen how the nuclear industry in the West has promised a lot, and at the beginning of a project the first thing to do is negotiate who will pay for cost overruns such that private investors are protected and added costs are covered by the state or the ratepayer.
Meanwhile, our construction companies are less experienced, our labor costs more, and our supply chains are all over the map.
C’mon, guys – this is bullshit! I can’t believe I’m the only one who’s noticed that the emperor has no clothes on. Nuclear new-build is economic and environmental nonsense. It’s a con! Tax credits, federal price guarantees and any other tricks that will “lower overall financing costs” – and they have the gall to claim it will befit the customer. Nuclear new-build is eye-wateringly expensive and entirely unproven; its construction delays are infinitely elastic and its pay-back time is pure fiction extending several decades into a unpredictable future. Irrefutable climate science tell us we need cheap, reliable, scalable means to produce carbon-free energy; and we need it now – not ten, fifteen, twenty years down the road. Renewables fit the bill. We don’t need nuclear!
I’m reminded of a TV ad from the 1990s. The prince rocks up at the palace and proudly parks his brand-new Peugeot 205 among the limos. Papa sheik looks down his nose and sneers, “Not expensive enough, son.”
Higher interest rates are a further death knell for such long, capital-intensive projects.
China has the advantage of a fiat government, a giant pool of relevant engineers, coherent investment, high PPP, increasing cumulative project experience and opaque safety guidelines.