Between 2020 and 2025, the power density of AI servers increased elevenfold. An individual server rack in an advanced AI data centre — roughly the size of a large refrigerator — could by 2027 draw peak power equivalent to 65. AC busway 480V ac PSU Power Distribution Unit (PDU) PSU IT Rack Power Supply 1ph AC Unit (PSU) VRs and PoLs 12V Compute/ comms chips <1V UPS Line Freq. Transformer 13kV ac grid Generators ATS/ STS Figure 2. First-generation traditional rack servers Figure 3 shows the next evolution in the data. The IEA's latest report, Key Questions on Energy and AI (April 2026), puts the updated trajectory plainly: consumption will roughly double and reach almost 500 TWh in 2025 to 950 TWh by 2030, with AI-specific infrastructure tripling over the same period. Five technology companies now spend more on. Today, the real constraint for scaling artificial intelligence is electricity—but in two distinct ways. Second, power density: once power reaches the facility. Understanding the role of data centres as actors in the energy system first requires an understanding of their component parts. Data centres are facilities used to house servers, storage systems, networking equipment and associated components that are installed in racks and organised into rows. Today, a single NVIDIA GB200 NVL72 AI rack draws 132 kW — more than 16 times as much. By 2028, racks are projected to reach 1 MW. 5% of total global electricity.
