- The International Energy Agency predicts that electricity demand will double and AI’s energy use could quadruple within the decade—but that may end up helping the planet.
- The report argues that if properly harnessed, AI could enhance energy efficiency across sectors, ultimately reducing net emissions.
The International Energy Agency predicts that electricity demand will double and AI’s energy use could quadruple within the decade—but that may end up helping the planet.
The surge in demand for AI could reshape global energy systems, with data centres poised to consume nearly as much electricity by 2030 as the entire nation of Japan, according to a new report from the International Energy Agency (IEA).
Driven by the insatiable energy appetite of AI technologies, data centre electricity demand is expected to more than double within the decade. AI-specific infrastructure will account for the lion’s share of this growth, with demand from AI data centres alone forecast to quadruple.
The IEA report highlights the potential for a transformation in how global power is consumed.
In the United States, electricity use from data processing, including AI, is projected to outstrip that of the entire manufacturing sector – which includes steel, cement and chemical production – by 2030.
A single large-scale data centre can use as much electricity as 100,000 homes, and some of the facilities now under construction are projected to be as much as 20 times larger.
Despite the scale of projected energy use, the IEA maintains that the threat AI poses to climate efforts has been “overstated.”
The report argues that AI, if properly harnessed, could enhance energy efficiency across sectors, ultimately reducing net emissions.
AI could be key in redesigning electricity grids to accommodate renewable sources like wind and solar, which produce power intermittently. Smart grids managed by AI might better balance supply and demand, replacing the rigid, fossil fuel-dependent legacy networks.
However, the report acknowledges that without strong regulatory frameworks and strategic planning, AI’s energy demands could spiral out of control. The expanding industry could offset hard-won energy efficiency gains and push utilities toward dirtier, more accessible energy sources.
Claude Turmes, a former Green MEP and Luxembourg’s ex-energy minister dismissed the IEA’s conclusions as overly optimistic and politically expedient.
“Instead of making practical recommendations to governments on how to regulate and thus minimise the huge negative impact of AI and new mega data centres on the energy system, the IEA and its [chief] Fatih Birol are making a welcome gift to the new Trump administration and the tech companies which sponsored this new US government,” he said.
Tom Winstanley, CTO & head of new ventures at NTT DATA UK&I, said, “It is clear that this growth trajectory of data centre energy usage cannot continue unchecked. As AI models continue to grow in scale and sophistication, so too must our data centre infrastructure strategies.
“Rather than simply expanding power capacity, we have an opportunity to innovate and reimagine how digital infrastructure is built and operates.”
The IEA’s predictions come as Donald Trump’s trade war threatens to derail the rapid expansion of large-scale battery storage across the US.