Will nuclear energy fuel the growing rise of artificial intelligence?

Global demand for electricity is growing at an accelerated rate. By 2035, it is expected to increase by more than 10,000 terawatt hours, equivalent to the total consumption of all advanced economies today.

The rise of artificial intelligence plays a big role in this, as AI technology relies on data centers, and an average-sized data center consumes as much electricity as 100,000 homes.

According to the International Energy Agency, demand for data centers increased by more than three-quarters between 2023 and 2024, and is expected to account for more than 20% of electricity demand growth in advanced economies by 2030.

In the USA, where many leading AI companies are based, energy consumption for AI-powered data processing is expected to exceed the total electricity consumption for producing aluminum, steel, cement and chemicals combined by the end of the decade.

Engine of the future

In December 2025, policymakers, technology companies and nuclear industry leaders from around the world gathered at IAEA headquarters in Vienna to explore the opportunities for nuclear energy to enable the expansion of artificial intelligence, and how artificial intelligence can drive innovation in the nuclear industry.

Training sophisticated AI models requires tens of thousands of CPUs to run continuously for weeks or even months. At the same time, the daily application of artificial intelligence is expanding to include almost all sectors, such as hospitals, public administration, transportation, agriculture, logistics, and education. Every query, every simulation, every recommendation consumes energy.

Manuel Greisinger, a director at Google who specializes in artificial intelligence, said: “We need clean, stable, carbon-free electricity available around the clock. This is undoubtedly a very high standard, and cannot be achieved using wind and solar energy alone. Artificial intelligence is the engine of the future, but an engine without fuel is almost useless. Nuclear energy is not just an option, it is an essential and indispensable component of the energy structure of the future.”.

Activity in the nuclear energy industry

In turn, IAEA Director General Rafael Grossi believes that the nuclear energy industry will be the energy partner for the artificial intelligence revolution.

Grossi said: “Nuclear power alone can meet the five needs of low-carbon power generation, round-the-clock reliability, ultra-high power density, grid stability, and true scalability.”.

There appears to be an increase in the nuclear energy industry, as 71 new reactors are being built, in addition to the 441 reactors currently operating worldwide. Ten reactors are scheduled to be built in the United States, which already has 94 plants, the largest number of any country.

Giant technology companies that use data centers have pledged to support the goal of at least tripling global nuclear power capacity by 2050. For example, Microsoft signed a 20-year power purchase agreement that allowed the first unit of the Three Mile Island nuclear power plant in Pennsylvania to be restarted.

Investments in Europe, Japan and the Middle East

The rest of the world is actively investing in nuclear energy, driven by the growth of artificial intelligence. The Director General of the International Atomic Energy Agency explained that “Europe has the densest digital corridors in the world, with Frankfurt, Amsterdam and London being major hubs.”.

He added: “Leading nuclear energy countries, such as France and the United Kingdom, are doubling their efforts in building nuclear power plants, and emerging countries such as Poland are accelerating their participation in this field.”.

With its distinguished research base in mathematics and computer science, Russia remains the world’s largest exporter of nuclear energy and is a leader in the operation and development of advanced reactor technology, while China is making significant advances in both artificial intelligence and nuclear energy.

In turn, Japan is intensifying the construction and modernization of data centers to meet the increasing demand. In the Middle East, the United Arab Emirates has established a nuclear energy program and has emerged as a regional hub for artificial intelligence.

Are small reactors the solution?

The need for more energy, and sooner, also drives the construction of small modular reactors, which differ significantly from traditional large power plants that require huge investments and a lead time of about 10 years.

Grossi said: “These types of reactors are characterized by their small size and advanced safety systems, and can be deployed in nearby industrial areas, including data center complexes.”.

Technology companies using these reactors don’t have to worry about regional grid supply constraints or loss of power during transportation. This will be a critical advantage in areas where network updating is slow, and interconnection queues are long.

Although this type of reactor still needs to get past the research and development stage, the IAEA is working closely with regulators and industry to make it a practical option, and we may soon see the deployment of large numbers of small reactors to meet demand.

For example, Google signed an agreement with an energy company to purchase nuclear power from several small modular reactors, a world first. If all goes well, these reactors could start operating by 2030.

Google is also turning its attention to space, where it is exploring space solar networks to enable large-scale machine learning in orbit, taking full advantage of solar energy. Two experimental satellites are scheduled to be launched in early 2027 to test radiation tolerance and data processing capabilities in the space environment.

Whether it is harnessing solar energy in space, restarting old reactors, investing in a new generation of small modular reactors, or building large reactors, all actions point in the same direction; Building an energy system based largely on nuclear energy that can meet the needs of future civilizations.