Global electricity demand for data centres is set to surge 26.4 per cent in 2026, with consumption climbing to 565 terawatt hours (TWh) — up from 447 TWh in 2025, according to a new report by technology research firm Gartner, Inc. released on Wednesday, 10 June 2025. The primary driver is compute-intensive artificial intelligence (AI) workloads pushing power needs to unprecedented levels.

Key Figures and Projections

Data centre power demand is forecast to reach 132 gigawatts (GW) in 2026, rising from 104 GW in 2025. The trajectory steepens sharply thereafter, with Gartner projecting demand to hit 290 GW by 2030 — nearly tripling in five years. Total electricity consumption for data centres is estimated to exceed 1,200 TWh by 2030, a scale that analysts warn will outstrip available grid supply.

Cooling systems and other supporting infrastructure are also under strain, with associated power needs rising 22.6 per cent in 2026 and a further 24.6 per cent in 2027.

AI Servers at the Centre of the Power Crisis

AI-optimised servers are the single largest contributor to the demand spike. According to Gartner, such servers will account for 31 per cent of total data centre power consumption in 2026. By 2027, their power draw is projected to surpass that of conventional servers — a crossover point that underscores how rapidly the AI hardware build-out is reshaping global energy infrastructure.

Linglan Wang, Director Analyst at Gartner, said: 'AI capacity is now constrained by power availability, making data centre power security the new battleground for scaling and protecting margins in the global AI race.'

Grid Supply Cannot Keep Pace

With projected consumption exceeding 1,200 TWh by 2030, existing grid infrastructure will be insufficient to support future data centre construction, according to the Gartner report. The shortfall is expected to affect all categories of data centre users — from hyperscalers to enterprise operators.

This comes amid a broader reckoning over the environmental footprint of digital infrastructure. A separate recent report found that data centres collectively consumed as much electricity as Saudi Arabia in the previous year. Should electricity use double by 2030, the associated carbon footprint would require the equivalent of 6.7 billion trees grown over a decade to offset.

What Infrastructure Leaders Must Do

Gartner is urging infrastructure and operations leaders to act now. Wang added: 'Infrastructure and operations (I&O) leaders must prioritise efficiency upgrades and secure grid access. They also need to invest in high-efficiency cooling systems and edge computing to mitigate power constraints and ensure sustainable, scalable growth.'

Notably, the convergence of power scarcity and AI ambition is reshaping investment priorities across the technology sector — with energy procurement and grid access increasingly treated as strategic assets rather than operational overhead. How governments and utilities respond to this demand curve will significantly influence where the next generation of AI infrastructure gets built.