CHINA AI2X BRIEFING

How AI is reshaping China’s Industries


When the Atom Meets the Algorithm

China deploys artificial intelligence in nuclear reactors that in turn power the country's Data Centres

Published on Dec 31, 2025

shared.image.missing_image

China recently won praise for its dual progress in artificial intelligence and nuclear energy from Rafael Mariano Grossi, the Director General of the International Atomic Energy Agency (IAEA). Speaking at the first-ever "International Symposium on AI and Nuclear Energy" in Vienna at the beginning of December, Grossi highlighted the country’s achievements along both trajectories.

“China is advancing on two fronts — artificial intelligence and nuclear energy — and has achieved remarkable results. The rapid development of its AI technologies is progressing in parallel with the construction of AI data centers, while at the same time China leads the world in the number of newly built nuclear reactors.”

The government in Beijing is indeed betting big on the new convergence of algorithms and atoms. China currently has 58 nuclear reactors in operation with a total installed capacity of 60.9 gigawatts, ranking second only to the United States. A further 28 nuclear reactors with a combined capacity of 33.7 GW are under construction, which amounts to around 50 percent of the total capacity under construction in the world.

Artificial intelligence is not only powered by nuclear energy, it’s also improving it.

Rafael Mariano Grossi, the Director General of IAEA

Now, AI is entering a phase of rapid deployment in China's reactors. Recently, China General Nuclear Power Group (CGN), one of China’s two dominant nuclear operators, released its new "Next-Generation Intelligent Industrial Control System (iNICS)".

It is an upgrade of its existing digital instrumentation and control system. CGN claims that it amounts to a "digital brain" integrating digital twin and AI technologies with a "closed-loop process of perception-analysis-decision-execution".

This is just one of many examples of the deployment of AI along the whole value chain of China's nuclear industry that is currently taking shape. At the same time, the central planners in Beijing have started to map out a future where nuclear energy will help to power the rising number of data centres the country needs to train large models and integrate AI into its industries.

The latest 2025 edition of the “China Nuclear Energy Development Report” explicitly acknowledges this trend of a two-way convergence between AI and nuclear energy. "On the one hand, the nuclear power industry is actively advancing digital and intelligent upgrades, applying AI technologies to engineering design and R&D, project construction, intelligent operations and maintenance, and intelligent decision-making. On the other hand, nuclear energy as a low-carbon, clean, safe, and stable energy, is increasingly favored by the AI computing industry as a reliable power source", China's official news agency Xinhua is quoting from the report.

Nuclear energy companies in China are currently teaming up with "leading information technology companies" to explore the use of Small Modular Reactors (SMR) to supply power to data centres, Xinhua also reported (in Chinese). The first SMR, named "Linglong-1", is scheduled to start commercial operations on China's southern island of Hainan next year.

Described as the world’s first commercial small modular reactor and sometimes nicknamed a “nuclear energy bank”, this small, modular type of reactor can be built faster than huge nuclear power plants, making them suitable for deployment where data centres are scaling up.

This vision of nuclear power supporting the growth of AI is currently gaining ground around the world, from from policymakers in Beijing to the boardrooms of Google, Meta or Dow Chemical.

“We are building the engine of the 21st century, artificial intelligence. But an engine without fuel is almost useless. Nuclear power is not just one option among many; it is an indispensable core component of the future energy mix,” the UN News Service quoted Manuel Greisinger, Director of Distributed Cloud at Google.

Greisinger explains why he believes that AI and nuclear energy are a particularly good match. “The power loads required by artificial intelligence are fundamentally different from the standard cloud computing workloads people are familiar with. They involve high power density, are extremely latency-sensitive, and require absolute continuity.”

The US administration plans to quadruple the country’s nuclear power capacity by 2050. It is also talking about developing a new generation of reactor technology especially for data centre needs.

While these discussions continue, China has begun to move toward implementation. Feasibility studies for the use of small modular reactors have begun.

These Chinese plans align with the vision of the IAEA. "We have a huge opportunity to make sure our digital future runs on clean energy. This is where Small Modular Reactors (SMRs) become especially relevant," says Grossi, the Director General. "They work particularly well for data centres because they are designed to be built in segmental units, making phased deployment possible. As an AI cluster expands, so can its nuclear power source."

Keep Reading

No posts found