Access to the Editor’s Digest has been made available at no cost. Roula Khalaf, the Editor of the Financial Times, compiles her preferred stories in a newsletter distributed weekly.
Recently, major technology companies have been making significant moves in the nuclear energy sector. Amazon has entered into agreements with utilities in Washington State and Virginia to support the development of small modular reactors (SMRs) and has also invested in X-energy, a company specializing in these reactors. Google has committed to purchasing power from SMRs that are to be constructed by Kairos Power, a start-up in the field. Additionally, Microsoft has entered into a 20-year power purchasing agreement with Constellation Energy, which will involve reopening a unit at the Three Mile Island plant in Pennsylvania, which had been closed since 2019.
The technology sector’s interest in nuclear energy is partially driven by the increasing energy demands associated with artificial intelligence. According to estimates from Goldman Sachs, energy consumption by data centers could rise by 160 percent by the year 2030. In the United States, the growing data needs are expected to significantly contribute to electricity demand, alongside the electrification of transport and a resurgence in manufacturing. Similarly, in Europe, power demand is projected to increase by 40 percent from 2023 to 2033. The International Energy Agency recently announced that the world is transitioning into the “age of electricity,” following the eras dominated by coal and oil.
Tech companies are aware that securing sufficient power for data centers, especially in countries like the United States, necessitates proactively arranging their own energy sources. The commitment to achieving net-zero carbon emissions requires them to invest in green energy, including wind and solar. Incorporating nuclear energy into their portfolios presents both an opportunity and a risk. Although nuclear energy has the potential to play a significant role in climate solutions due to its low carbon emissions and reliability, the construction of large nuclear plants is often prohibitively expensive and time-consuming. SMRs, which range up to 300 megawatts compared to 1,000 megawatts for traditional nuclear plants, offer a potentially more cost-effective and quicker alternative. These reactors, which are mostly prefabricated, can be installed close to areas with high power demand and on sites like decommissioned coal plants already connected to the grid.
However, there are challenges associated with SMRs, including possibly high costs for regulatory approvals in a sector where safety is crucial. There are also concerns that they might divert necessary investment from established solar, wind, and battery power systems. Currently, SMRs remain relatively untested, and the Institute for Energy Economics and Financial Analysis has deemed them “still too expensive, too slow, and too risky” based on existing projects.
The involvement of major technology companies, with their substantial financial resources and innovative capabilities, might accelerate the development of SMRs and shift the focus from government-led nuclear projects to private sector initiatives. An example of this shift is the impact Elon Musk has had on the space industry. Nevertheless, reopening or extending the lifespan of existing nuclear facilities might be a more practical approach. Besides Three Mile Island, a nuclear plant in Michigan is also being recommissioned.
Regardless of the approach, the anticipated increase in AI-driven data demand before 2030 suggests that technology companies will need to continue investing in wind and solar energy. In the competitive landscape for resources, regulators must ensure that tech companies with substantial financial resources do not monopolize new energy supplies. One possible regulation could require that clean energy projects dedicated to data centers also be scaled to supply the grid or other consumers. Furthermore, artificial intelligence could be leveraged to enhance energy efficiency in various sectors including factories, offices, and across power grids. In this new age of electricity, AI should not merely be an additional consumer of energy but should be central to achieving green energy solutions.
