Global electricity demand from data centers could rise from about 485 terawatt-hours in 2025 to roughly 950 terawatt-hours in 2030, according to the International Energy Agency.
The forecast shows how quickly AI is becoming an energy-system issue. The IEA estimates that technology companies invested more than $400 billion in 2025 and expected capital spending to grow sharply again in 2026. Capacity in large AI-focused facilities has already expanded rapidly as companies race to secure compute.
AI workloads are especially important because their power requirements are concentrated in large campuses and high-density server racks. A national grid may have enough generation in aggregate while still facing serious local constraints around substations, transmission lines and the time needed to connect a new data center.
This creates opportunities for utilities, grid-equipment suppliers, storage providers and developers of nuclear and renewable power. It also makes efficiency more valuable: improvements in chips, cooling and workload scheduling can reduce the amount of new infrastructure required for each unit of AI output.
Source: International Energy Agency
