Big Tech companies are making significant moves toward nuclear energy. Recently, Amazon partnered with utility companies in Washington state to work on developing four small modular reactors (SMRs). Virginia saw a similar agreement, and Amazon also invested in X-energy, a company focusing on SMR technology. Google has plans to purchase power generated from SMRs that will be developed by a start-up called Kairos Power. Additionally, Microsoft secured a 20-year agreement with Constellation Energy, allowing for the reopening of a unit at the Three Mile Island plant in Pennsylvania, which had been closed since 2019.
The tech sector’s interest in nuclear power is partly driven by the high energy demands of artificial intelligence (AI). Running an AI query can consume as much as ten times the energy required for a regular Google search. According to Goldman Sachs, electricity demand from data centers is projected to increase by 160% by 2030. Factors like increased data needs, electrifying transportation, and efforts to bring back manufacturing jobs are expected to boost electricity demand growth in the U.S.
In Europe, Goldman Sachs estimates that power demand could rise by 40% from 2023 to 2033. The International Energy Agency announced that the world is entering a new era focused on electricity, following the ages of coal and oil.
Tech companies understand that if they want to build data centers in places like the U.S., they need to secure a reliable energy supply. Their commitments to achieve net-zero emissions demand that the energy sources be renewable. Having already invested substantially in wind and solar, the shift to include nuclear power makes sense but comes with uncertainties.
Nuclear energy has the potential to contribute to climate solutions since it generates low carbon emissions and can provide reliable power for long periods. However, constructing large nuclear power plants can be prohibitively expensive and take a long time to build. In contrast, SMRs, which can generate up to 300 megawatts, propose a quicker and more cost-effective approach. These smaller reactors can be pre-fabricated and installed at locations where energy is most needed, such as old coal plants.
Despite their advantages, SMRs may also face steep costs relating to design approval by regulators, especially considering safety is a top priority. There are concerns these initiatives could divert financial resources from established renewable energy options like solar and wind energy. Current evaluations of SMR projects indicate that they remain untested and have high costs associated with them.
However, the financial strength and innovative approaches from big tech might help advance SMRs, promoting a transition from government-led nuclear projects to privately financed initiatives. Past examples show how entrepreneurial efforts have revolutionized sectors, as seen with Elon Musk in the space industry. Reopening existing nuclear facilities, like those in Three Mile Island and Michigan, may be a more pragmatic solution.
Regardless, the increasing data requirements driven by AI suggest that major tech companies will need to continue investing in renewable energy sources. As these companies vie for energy resources, regulators must ensure that they do not monopolize new energy supplies. One potential solution could involve mandating that clean energy projects designed for data centers also supply the broader grid or other users. There’s also potential for using AI to enhance energy efficiency in various sectors. As we step into this new age of electricity, it’s essential that AI not only consumes energy but also plays a vital role in contributing to greener solutions.