Why the Nuclear Renaissance Keeps Falling Behind Schedule

Demand for electricity is quickly outpacing how fast new supply can come online. This increased demand is felt on energy bills in the form of high capacity charges. 

The strain on the grid is part of the reason there is a push in the United States for a nuclear renaissance. Nuclear power is reliable and clean, however building new plants can be a long process that is very costly. 

Why would nuclear power actually help with grid strain?

Nuclear plants run at a capacity factor of about 92 percent, the share of time a plant produces power at its full potential, the highest of any energy source. That reliability comes from needing to refuel only every 18 to 24 months instead of cycling on and off with the weather or the market. This supply of uninterrupted power is crucial with data centers running artificial intelligence (AI) workloads around the clock, placing more demand on the grid. Nuclear power already makes up close to a fifth of U.S. electricity generation, according to the U.S. Energy Information Administration. Nuclear is one of the few sources built to run at that scale without interruption.

So why is the nuclear renaissance still stuck at the starting line?

Georgia's Plant Vogtle is the clearest example. The two new reactors were originally budgeted at $14 billion and expected to begin commercial operation around 2016 and 2017. Construction started in 2009, but the EIA now reports total project costs above $30 billion, and the second reactor did not reach commercial operation until 2024, 15 years after construction began. 

Part of the delay is regulatory. The Nuclear Regulatory Commission (NRC) requires safety and environmental reviews that routinely stretch for years, on top of the construction time itself. Fuel supply adds another constraint: newer reactor designs depend on enriched uranium capacity that the domestic supply chain has not caught up to yet, and a federal phase-out of Russian enriched-uranium imports adds further pressure on that timeline. 

Restarts should be the fast path because the infrastructure already exists, but even restarts are running long. Constellation's Crane Clean Energy Center (formerly Three Mile Island) is moving through NRC environmental review, but a final decision on its operating license isn't expected until May 2027, three years after Constellation announced the restart plan in 2024. In Michigan, Holtec's restart of the Palisades plant, backed by $1.52 billion in DOE loan guarantees, has already slipped past its original timeline amid corrosion and supply-chain issues.

Once a plant is built, a finished reactor does not automatically deliver power. New generation, nuclear included, has to clear a grid operator's interconnection queue, the review process that determines whether a plant can actually connect to transmission lines, before it counts toward capacity. A Utility Dive report on PJM's interconnection backlog found that most projects entering the queue today are unlikely to come online before 2030, regardless of how quickly construction itself finishes.

Do small modular reactors change the timeline?

Small modular reactors (SMRs) are scaled-down nuclear reactors, typically under 300 megawatts versus roughly 1,000 megawatts for a conventional plant, designed with standardized parts that can be factory-built and shipped to a site rather than custom-engineered on location. 

Although SMRs could offer a quick solution to bring nuclear power online, they are not without their challenges. NuScale, the first SMR designer to win NRC licensing (2023), saw its flagship project, seven small reactors for a 462-megawatt plant in Idaho, collapse after cost estimates more than doubled, from under $4 billion to over $9 billion, before a single unit was built. Holtec, the same company restarting Palisades, is racing to build two SMRs next door on that same site; its own spokesman put the realistic timeline at two and a half more years of NRC licensing plus two and a half years of construction, assuming no setbacks, on top of 15 years of design work already invested. Worldwide, only three SMRs are running today: two 35-megawatt units in Russia and one 125-megawatt unit in China. None operate yet in North America or Europe.

No U.S. SMR design has been built commercially yet, so the promise of a shorter, more standardized build has yet to be delivered. Businesses planning around nuclear power, new reactors or small ones, should treat the nuclear renaissance as a potentially multi-decade story rather than a multi-year one. 

Key Takeaway

Nuclear power is reliable enough to help ease grid capacity strain, but the nuclear renaissance is unfolding on a decade-plus timeline, not a multi-year one. Cost overruns, regulatory review, fuel supply limits and grid interconnection queues all push completion dates further out than early announcements suggest. Businesses managing energy costs today should plan around existing capacity constraints rather than count on new nuclear supply arriving soon.