Data centres consume more electricity than most countries
International Energy Agency data show data centres used about 415 TWh of electricity in 2024, roughly 1.5% of global consumption. By 2030 the figure could reach 945 TWh.
The rapid growth of artificial intelligence is putting increasing strain on the world's energy supply. Data from the International Energy Agency (IEA) indicate that data centres worldwide consumed about 415 terawatt-hours (TWh) of electricity in 2024, around 1.5% of all electricity used globally.
If data centres were counted as a single country, they would rank among the eleven largest electricity consumers, with only ten nations using more. Forecasts suggest consumption will more than double by 2030 to 945 TWh, close to Japan's current annual use.
Concentrated in a few regions
About 85% of the energy used by data centres is consumed in three regions: the United States, China and Europe. While 1.5% may look small globally, the local effect is greater. In regions with many data centres, they can take 25–30% of total grid capacity.
In Ireland, data centres began using more electricity than all of the country's urban households combined, and the government introduced strict limits on constructing new facilities.
Environmental burden
Running data centres is estimated to cause about 180 million tonnes of indirect carbon dioxide emissions. Generating electricity for them and cooling them could require roughly 9.3 trillion litres of water by 2030. Technology companies are seeking solutions in renewable energy and nuclear power, but fossil fuels still play a significant role in many regions.
Resistance to new projects
Concerns about grid overload, rising tariffs and depleted water resources are already slowing the development of new facilities. Because of opposition from the public and local authorities, investments of about €37.5 billion have been halted or delayed in Europe alone. At least 15 US states are considering moratoriums on new construction. The main challenge ahead is to balance the need for powerful AI infrastructure with the quality of life of local communities.
