Between late 2016 and August 2026, China connected 27 new nuclear reactors to its grid. In the same period, the United States connected two. As Jacopo Buongiorno—professor of nuclear science and engineering and director of the Center for Advanced Nuclear Energy Systems at the Massachusetts Institute of Technology—tells me, “China is the world’s leader in nuclear power plant (NPP) construction.”
It’s an impressive data point on its own but when you consider the greater picture, it’s even more striking. Those nuclear power plants are only a small fraction of China’s current power throughput, but they are the crux of the country’s electricity generation. The reason is that Chinese engineers have designed a power grid that relies heavily on renewables (and aims to eventually be 100% green). But, since wind and solar generation fluctuate a lot, they need a lot of nuclear generation to keep their electrical network stable and fully operational. Without that constant power, the fluctuations will make the grid crash.
Beijing’s electricity hunger traces back to Deng Xiaoping’s “Reform and Opening-Up” policies, launched in December 1978, which set China on a path of rapid industrialization, market liberalization, and urbanization that has continued for nearly five decades. That transformation drove a surge in electrification, originally fed by coal and, in recent years, increasingly by renewable energy.
“We’re in this incredible race to add enough power generation capacity in China to meet what is adding one extra Germany of new electricity consumption every year,” David Fishman, a principal at energy consultancy firm The Lantau Group, recently told the South China Morning Post.
The math checks out: China’s power consumption surpassed 10 trillion kilowatt-hours—10,000 terawatt-hours (TWh)—in 2025, pushed up by advanced manufacturing and artificial intelligence. China’s electricity demand grew by 503 TWh that year alone. Germany’s gross electricity consumption, from every source combined, was estimated at roughly 517 TWh over the same period—meaning China added nearly one Germany’s worth of new demand in a single year.
Almost none of that growth is being met with nuclear power. Coal still supplies the largest share of China’s electricity, but renewables are closing the gap fast: Wind and solar rose to 22% of national generation in 2025, up from 18% the year before, and by the first half of 2026, coal’s share had fallen below 50% of total generation for the first time on record. Nuclear, meanwhile, holds steady at about 5% of the mix.
If the extreme electricity build-up is almost entirely driven by wind and solar now, why is nuclear so important for China’s power grid design? Designing a grid is not a matter of building power plants and connecting them to cities and factories the same way you connect your toaster to the wall plug. Energy production fluctuates, and every node in a national grid needs to be controlled and balanced so the entire network doesn’t go down (see what happened during the Spanish power outage last year, where renewables produce most of the energy now, but it’s not well balanced).
To avoid a disastrous Spanish situation, China has been doing two things for close to two decades now. First, building a “bullet train for power”—the world’s most extensive ultra-high-voltage grid, which carries electricity from wind and solar farms in the west to industrial hubs on the coast, balancing supply and demand across the country in real time. Second, building out a nuclear fleet that supplies power regardless of weather, wind speed, or time of day. That’s where nuclear energy comes in.
The atomic key
Nuclear supplies something called continuous baseload output to the power grid. A grid built overwhelmingly on intermittent generation needs a fixed floor under it, and nuclear is the floor Beijing has chosen in its race to zero-carbon power.
China currently operates 64 commercial reactors producing 63.9 gigawatts, the second-largest fleet in the world behind the U.S. Domestic companies are building 37 more units totaling 38.5 gigawatts. The State Council has approved 10 or more reactors annually since 2022, including eight units across four projects greenlit on July 31, 2026, worth $25 billion, on the way to 110 gigawatts of installed nuclear capacity by 2030 and a power mix that is 50% non-fossil.
The astonishing nuclear reactor construction speed comes first from standardization. China builds two pressurized water reactor designs almost exclusively: the domestic third-generation Hualong One, and the CAP1000, adapted from an American design created by Westinghouse.
“The supply chain is all built around this standardization,” Victor Nian of the Centre for Strategic Energy and Resources told the South China Morning Post. Domestic suppliers now manufacture nearly every component a new reactor needs. That repetition compresses build time to about five years per unit, while in the West we spend a decade or more.
You may think that, perhaps, China is skipping safety steps, but the fact is that they are not, Buongiorno tells me. The country’s existing fleet, essentially all water-cooled reactors, already operates safely and reliably, he says, and, “there is no reason to believe the NPPs being built now, which are essentially identical, will perform any differently.” He says that China, like every country that operates nuclear power plants, belongs to the World Association of Nuclear Operators, which conducts periodic reviews and inspections of member fleets. “China does not run these NPPs in isolation,” he points out.
The other element behind this speed is state financing, which compresses the cost curve the same way standardization compresses the timeline. According to the SCMP, developers borrow at interest rates as low as 1.5%, bringing the levelized cost of electricity for a 1-gigawatt reactor to roughly $42 per megawatt-hour, about half what a comparably financed Western project would cost.
Buongiorno frames this as a proven approach rather than an improvised one. “State-backed financing and stable energy policies are a good way to ensure these capital-intensive and long-duration projects can be carried out to completion with minimal perturbations,” he says, pointing to France’s nuclear build-out in the 1970s and 80s, and to India, Russia and South Korea today, as prior examples of the same model.
Beijing’s model has one major physical limit: China permits commercial reactors only on the coast, where suitable sites cap out at roughly 163 reactors. Beijing suspended approvals for inland reactors after the 2011 Fukushima meltdown. “China will physically run out of suitable sites at some point,” Fishman told the SCMP. Lifting that moratorium would open 77 sites nationwide with room for 345 reactors.
Buongiorno does not see the inland shift as a new category of risk, however. “Inland reactors are in operation in dozens of countries around the world,” he tells me. “Their risk profile depends on the specifics of the site—seismicity, water access, population density, etc. Nothing new here.”
