Google will buy half the electricity of a nuclear power plant

If you are building or operating large-scale computing infrastructure, the cost of electricity for data centers is now partially hedged by long-term nuclear power purchase agreements.

Google has announced a record €13 billion investment in Finland, focusing on expanding its AI infrastructure. This investment includes funding for three new data centers, an expansion of an existing site, and energy projects to meet the growing demand for AI services. A key component of this initiative is a 22-year contract with Finnish utility Fortum, through which Google will purchase up to half of the electricity generated by the Loviisa nuclear power plant. This deal, set to support over 37,000 jobs and boost Finland's GDP, underscores Finland's appeal for tech companies due to its cool climate, low-carbon energy, and robust digital infrastructure. The investment is slated to commence construction in 2027 and 2028, supporting Google's AI chatbot Gemini and other services like Search and Maps. Google's parent company, Alphabet, has previously signaled substantial global spending increases for AI computing capacity.

This strategic move by Google highlights the escalating demand for reliable and substantial power sources to fuel the AI revolution. By securing up to 50% of a nuclear power plant's output, Google is not only ensuring a stable energy supply but also demonstrating a commitment to the long-term operational costs associated with massive data center expansion. This precedent could influence other major tech players to explore similar long-term energy procurement strategies, potentially reshaping energy markets and accelerating investments in low-carbon power generation, particularly nuclear.

The pact between Google and Fortum offers long-term financial predictability for the Loviisa plant, enabling an investment program to extend its lifespan and increase its capacity. This demonstrates a symbiotic relationship where tech giants secure critical resources, while established energy providers gain the stability needed for essential upgrades and future planning. As AI's energy appetite grows, such large-scale energy-infrastructure collaborations could become increasingly common, potentially impacting grid stability, energy pricing, and the pace of decarbonization efforts globally.