According to the latest research from the International Energy Agency (IEA), global data center electricity demand is projected to double by 2030, surpassing Japan's total electricity consumption, with artificial intelligence (AI) being the primary driver. The IEA report indicates that data centers are expected to fuel over 20% of electricity demand growth in developed economies within the next five years, with the majority of this growth attributed to the widespread adoption of AI.
IEA statistics show that in 2024, the United States accounted for 45% of global data center electricity consumption. By 2030, this percentage is projected to rise significantly, with US data center electricity consumption expected to exceed that of the nation's entire energy-intensive manufacturing sector, including aluminum, steel, cement, and chemical industries. In the US, data centers will account for nearly half of electricity demand growth over the next five years.
Fatih Birol, IEA Executive Director, stated: "With the rise of AI, the energy sector is at the forefront of the most important technological revolution of our time." The report also highlights the need for increased renewable energy generation capacity to meet growing demand, while expressing optimism about AI's potential to optimize systems and reduce emissions.
The report mentions that AI is already being used in the energy sector to improve power generation, transmission, and consumption, and to help the oil and gas industry optimize exploration and production. The IEA believes that policy and regulatory shifts will be key to unlocking AI's potential benefits. However, concerns that AI might exacerbate climate change are deemed overstated by the IEA.
Although data centers are one of the fastest-growing sources of emissions, according to the IEA's worst-case scenario, their share of energy sector emissions will remain below 1.5% by 2035. However, the IEA doesn't expect AI to solve climate change entirely. The report notes that emission reductions from the widespread use of existing AI applications might only equate to about 5% of global energy-related emissions, which is far from sufficient to address climate change.
It's noteworthy that the energy use and emissions associated with chip production are equally significant. According to Greenpeace, global AI chip manufacturing electricity consumption surged over 350% between 2023 and 2024 and could reach 170 times its 2023 level by 2030, potentially exceeding Ireland's current electricity consumption.
Greenpeace expresses concern that in East Asia, most chip manufacturers are relying on fossil fuels to meet their growing energy needs. The IEA points out that despite the energy intensity of AI hardware manufacturing, its energy consumption remains lower than operational consumption over its product lifecycle. This contrasts sharply with the current trend of surging energy use in chip production.
Key Takeaways:
🌍 Global data center electricity demand is projected to double by 2030, primarily driven by AI.
💡 US data centers will account for nearly half of electricity demand growth over the next five years.
🔋 Electricity consumption for chip production is surging, potentially surpassing Ireland's electricity consumption.