English
Data centers around the world will double their electricity consumption by 2030, driven by the explosive growth in demand from the development of artificial intelligence (AI), according to a new report from the International Energy Agency (IEA). Intelligence, even artificial, comes at a cost!
In 2024, these centers consumed an estimated 415 TWh of energy - 1.5% of global consumption - but this figure is expected to reach 945 TWh by 2030, equivalent to the total electricity consumption of Japan today.
• Data centers, heavy consumers
According to the report, data centers are becoming increasingly energy-intensive, especially those supporting generative AI technology. AI algorithms require massive computing power and rapid access to vast databases, which leads to high electricity consumption. "A 100 MW data center consumes as much as 100,000 households per year. The largest ones under construction will consume 20 times that much - the equivalent of 2 million households," warns the IEA.
• US, Europe and China dominate energy map
The US, Europe and China are responsible for 85% of data center energy consumption, and the US is expected to generate almost half of the additional demand by 2030. The uneven distribution of these facilities - concentrated near major cities - creates challenges related to electricity infrastructure and the constant and secure supply of energy.
• Energy sources: mix of coal, gas and renewables
Coal currently accounts for 30% of data center energy consumption, but the IEA estimates that natural gas and renewables (solar, wind) will become dominant, due to lower costs and greater availability. States will adopt various strategies - for example, the US has launched a "National Energy Dominance Council" to support the energy production needed to compete globally in AI, especially against China.
• Climate impact: rising emissions, but with potential to offset
High consumption will also bring with it an increase in carbon dioxide emissions, from 180 million tons of CO₂ today to 300 million by 2035, the report says. Even though this value remains below 1.5% of global emissions from the energy sector, the growth rate is one of the fastest. "Fears that AI will accelerate climate change appear overstated, as do expectations that AI will solve the problem on its own," the report stresses. However, the IEA believes that AI could indirectly contribute to reducing emissions in other sectors, by making industrial processes more efficient, optimizing energy networks and reducing waste.
Euro
Dolar SUA
Franc elveţian
Liră sterlină
Gram de aur
