Technology / Economy

Add as a preferred source on Google

ADIPEC Survey Finds Executives Expect AI to Push Power Demand Higher

The National reported that 81% of respondents expect AI-related electricity use to exceed estimates; 32% expect it to be much higher.

Updated 2 min read
High-voltage transmission lines cross the outskirts of Abu Dhabi, United Arab Emirates. (AI-generated image / EconoTimes)
High-voltage transmission lines cross the outskirts of Abu Dhabi, United Arab Emirates. (AI-generated image / EconoTimes)

Artificial intelligence’s growing electricity needs are prompting energy executives to expect higher demand than current projections show, while the International Energy Agency says grid and equipment limits may restrain the buildout in the near term.

The National reported that 81% of respondents to ADIPEC’s survey expect AI-related electricity demand to exceed current estimates; 32% anticipate it will be much higher. ADIPEC’s public announcement describes the result as more than four in five, without stating the precise 81% figure. ADIPEC said its inaugural Global Energy Report 2026: Power Shift drew on responses from more than 1,800 senior business and energy leaders across 10 markets.

The survey captures respondents’ views, not a separate estimate of how much electricity AI will consume. For comparison, the IEA projects global data-center use at 485 terawatt-hours (TWh) in 2025 and 950 TWh in 2030, equal to about 3% of worldwide electricity demand that year. The agency expects AI-focused data-center consumption to reach three times its 2025 level by 2030.

Data-center electricity use increased 17% in 2025, the IEA said, with use at AI-focused centers rising 50%. Its central projection remains close to the outlook it published in 2025. Grid connections, planning and equipment supply chains limit how quickly new capacity can come online, though faster infrastructure construction and more energy-intensive AI applications could push demand higher after 2030.

The IEA said concentrated data-center loads may require investment in generation and grids. The IEA said prices could rise in some places if investment does not keep pace with demand. The agency also points to the need for generation and grid investment to serve concentrated data-center loads.

The MIT Center for Energy and Environmental Policy Research working paper Who Pays for Growth? Evidence from Data Centers and the Grid estimated that data-center entry raised U.S. household electricity prices by 2.1%. The finding was covered in EconoTimes’ earlier report; the estimate concerns the study’s U.S. analysis.

Survey respondents also see a shift in influence: 43% expect technology and AI companies to gain the most influence over global energy in the coming decade, and 71% view AI more as an opportunity than a risk, ADIPEC said. “AI is reshaping what the industry needs to deliver,” said Christopher Hudson, president of dmg events, the conference organizer.

ADIPEC 2026 is scheduled for Nov. 2-5 in Abu Dhabi.

Related articles

View all