Artificial intelligence has the potential to transform the energy sector in the coming decade, driving a surge in electricity demand from data centres globally while also unlocking significant opportunities to cut costs, enhance competitiveness and reduce emissions, according to a new report from the International Energy Agency (IEA).
Drawing from new datasets and wide-ranging input from policymakers, the tech industry, energy players, and global experts, the report forecasts that electricity use by data centres will more than double by 2030, reaching approximately 945 terawatt-hours (TWh)—slightly more than Japan’s total current electricity consumption.
“AI will be the most significant driver of this increase,” the report states, noting that electricity consumption from AI-optimised data centres alone is expected to quadruple by the end of the decade.
To meet this rising demand, a mix of energy sources will be used, though renewables and natural gas are anticipated to dominate due to their cost-effectiveness and availability in key markets.
The IEA report also pointed to uncertainties ahead—ranging from the global economic outlook and the pace of AI adoption, to how efficient AI systems will become and whether infrastructure bottlenecks in the energy sector can be resolved in time.
AI could intensify some energy security strains while helping to address others, according to the report. Cyberattacks on energy utilities have tripled in the past four years and become more sophisticated because of AI.
At the same time, AI is becoming a critical tool for energy companies to defend against such attacks. Another energy security concern relates to the expanding demand for critical minerals used in the equipment in the data centres that power AI.
As AI becomes increasingly integral to scientific discovery, the report finds that it could accelerate innovation in energy technologies such as batteries and solar PV.
“With the rise of AI, the energy sector is at the forefront of one of the most important technological revolutions of our time,” said IEA executive director Fatih Birol. “AI is a tool, potentially an incredibly powerful one, but it is up to us – our societies, governments and companies – how we use it.”
According to the report, countries that want to benefit from the potential of AI need to quickly accelerate new investments in electricity generation and grids, improve the efficiency and flexibility of data centres, and strengthen the dialogue between policy makers, the tech sector and the energy industry.
Goldman Sachs research forecasts global power demand from data centres will increase 50% by 2027 and by as much as 165% by the end of the decade.
On the demand side for data centres, large “hyperscale” cloud providers and other corporations are building large language models (LLMs) capable of natural language processing and understanding.
These models must be trained on vast amounts of information, using power-intensive processors.
On the supply side, hyperscale cloud companies, data centre operators, and asset managers are deploying large amounts of capital to build new high-capacity data centres.
At present, Goldman Sachs Research estimates the power usage by the global data centre market to be around 55 gigawatts (GW).
This is comprised of cloud computing workloads (54%), traditional workloads for typical business functions such as email or storage (32%), and AI (14%).
By modelling future demand for each of these workload types, the firm’s analysts project power demand will reach 84 GW by 2027, with AI growing to 27% of the overall market, cloud dropping to 50%, and traditional workloads falling to 23%.
This baseline scenario could, however, be affected by a deceleration in usage by AI — for example, if the transition to AI-driven work and AI monetization doesn’t develop as quickly as anticipated. In such muted scenarios, demand could diverge from the baseline estimate by 9-13 GW.
The current global market capacity of data centres is approximately 59 GW. Roughly 60% of this capacity is provided by hyperscale cloud providers and third-party wholesale data centre operators.
The remaining belongs to more traditional corporate and telecom-owned data centres.
The AI-dedicated data centre is an emerging class of infrastructure. Although very few exist so far, they’re designed for the unique properties of AI workloads — high absolute power requirements, higher power density racks, and the additional hardware (such as liquid cooling) that comes with it.
Goldman Sachs research estimates that there will be around 122 GW of data centre capacity online by the end of 2030. The mix of this capacity is expected to skew even further towards hyperscalers and wholesale operators (70% versus 60% today).
Given the higher processing workloads demanded by AI, the density of power use in data centres is likely to grow as well, from 162 kilowatts (kW) per square foot to 176 kW per square foot in 2027.
As data centres contribute to a growing need for power, the electric grid will require significant investment. Goldman Sachs Research estimates that about $720 billion of grid spending through 2030 may be needed.
