How AI is remaking and straining the global energy system

Published: 2026-07-05T12:20:38.000Z

The relationship between artificial intelligence and energy has entered a decisively new phase, and it now runs in both directions at once. AI is reshaping how power is generated, distributed and traded, while becoming one of the hungriest consumers of electricity ever built.

The relationship between artificial intelligence and energy has entered a decisively new phase, and it now runs in both directions at once. For years the story was mostly about efficiency, with clever algorithms trimming waste from grids and squeezing a little more from ageing infrastructure. That has given way to something far larger, as AI moves from tinkering at the edges to orchestrating how power is generated, distributed and traded, while simultaneously becoming one of the hungriest consumers of electricity the world has ever built. The technology is reshaping the energy system and straining it in the same breath. On the reshaping side, the tools have grown genuinely powerful. AI systems now forecast demand, balance the flicker of wind and solar, isolate faults in milliseconds and knit thousands of small batteries and rooftop panels into virtual power plants that can bid into wholesale markets like a conventional generator. A new class of energy trader is emerging around these models, and in the United States grid operators such as PJM have begun leaning on machine learning to speed up the notoriously slow process of connecting new projects. In oil and gas, the same techniques are being turned on exploration, drilling and production to cut downtime and lift output. The other side of the ledger is where the strain shows. The International Energy Agency reports that electricity demand from data centers climbed 17 per cent in 2025, with consumption from AI-focused facilities surging by half, and it expects data center demand to more than double by the end of the decade. The scale of spending is staggering, with the capital expenditure of just five technology companies now exceeding global investment in oil and gas production, and satellite tracking showing purpose-built AI factories have more than tripled their capacity in eighteen months. All of that competes for power, water and grid capacity, and ordinary consumers are already feeling it in higher bills. The Gulf has positioned itself as the proving ground for this whole tension, and its advantage is precisely energy. With electricity at roughly five to six cents a kilowatt-hour, well below American levels, Saudi Arabia and the UAE can fund compute at a scale others struggle to match, and they are doing exactly that. HUMAIN is targeting close to two gigawatts of AI data center capacity by 2030 with ambitions far beyond, while Abu Dhabi's G42 anchors a buildout that includes one of the largest AI campuses outside the United States. The region's data center market is set to nearly triple by 2030. Yet the same period exposed the fragility of this bet, after drone strikes on cloud facilities in the UAE and Bahrain turned questions of energy and resilience into matters of national security.