The increasing demand for electricity to power data centers running AI applications has led to a unique business opportunity for developers of small nuclear reactors and battery technologies. NuScale Power, a startup specializing in small modular nuclear systems, is one of the companies looking to profit from the energy-intensive needs of data centers. The company’s reactors could potentially operate independently of the grid, providing a reliable and carbon-free source of electricity for AI data centers. NuScale is working on a project with Standard Power to supply 24 SMRs capable of generating nearly 2 gigawatts of electricity.
The energy demands of AI data centers are significantly higher than traditional data centers, with AI-oriented facilities requiring about 80 megawatts of electricity. To meet these demands, companies are exploring different options, including small nuclear reactors and new battery technologies. For example, Nvidia’s H100 chip, a popular AI server box, uses up to 10 kilowatts of electricity, generating excess heat that requires additional cooling using up to 30% more water. Companies like Microsoft are also considering using SMRs to power their AI data centers, while startups like Radiant Industries are developing microreactors to provide additional energy as needed.
In addition to nuclear power systems, companies like Natron Energy are exploring sodium-ion batteries as a solution to the power surges associated with AI servers. These batteries can smooth out the distribution of electricity, preventing overloads and ensuring stability for the grid. While sodium batteries offer benefits like rapid recharging and the ability to handle sudden power surges, they are still relatively new and not widely available yet. However, with the increasing demand for energy solutions in AI data centers, companies like Natron Energy are seeing growing interest from customers.
Developing sustainable energy solutions for AI data centers is essential for meeting the growing demand for electricity in the tech industry. In addition to nuclear and battery technologies, companies are exploring other options like geothermal energy systems as a way to power data centers. Fervo is one startup focusing on geothermal energy, with Google testing their system in Nevada. While challenges remain in terms of safe storage of spent fuel rods and availability of clean fuels like green hydrogen, innovators are working on alternative solutions to meet the energy needs of AI data centers.
As companies continue to invest in energy solutions for AI data centers, the market for small nuclear reactors, batteries, and other technologies is expected to grow. These developments are crucial for ensuring a sustainable and reliable source of power for the growing number of data centers worldwide. With the support of investors and advancements in technology, companies like NuScale Power and Natron Energy are well-positioned to capitalize on the increasing demand for energy-efficient solutions in the AI industry. The future of AI data centers will rely on innovative technologies that can meet the high energy demands while also reducing carbon emissions and ensuring the stability of the grid for years to come.
