The International Energy Agency released "Energy and Artificial Intelligence."

Key Chapter Title List

1. The Rise of Artificial Intelligence and Its Connection to Energy
2. Energy Demand of Artificial Intelligence
3. Artificial Intelligence Empowering Energy Optimization
4. Artificial Intelligence Promoting Energy Innovation
5. Energy Security, Investment, and Policy Landscape
6. Power Consumption and Outlook of Data Centers
7. Power Supply to Meet Data Center Demand
8. Interaction Between Data Centers and the Power Grid
9. Application of Artificial Intelligence in Energy Systems
10. Application of Artificial Intelligence in Oil, Gas, and Critical Mineral Supply
11. Application of Artificial Intelligence in the Power Sector
12. Application of Artificial Intelligence in Energy End-Uses

Document Introduction

Artificial Intelligence (AI) is reshaping the global economy and industrial landscape at an unprecedented pace, and its rapid development has sparked widespread concern about its profound impact on energy systems. As the world's authoritative body in the energy field, the International Energy Agency (IEA) has released its first special report comprehensively examining the interaction between AI and energy. This report aims to fill existing knowledge gaps, providing the most comprehensive data-driven analysis to date, based on newly built global models, a comprehensive dataset on data center electricity demand, and extensive consultations with policymakers, the technology sector, the energy industry, and experts.

The report first examines the macro background of the AI technological revolution, noting that it has evolved from an academic field into an industry with a trillion-dollar market value, profoundly influencing corporate strategy, economic policy, and geopolitics. However, the development of AI is highly dependent on energy, as its training and operation occur in data centers with massive electricity consumption. The core analysis of the report revolves around two main themes: first, how much energy AI will require and how to meet this demand; second, how AI can be applied in the energy field to build a more secure, affordable, and sustainable energy system.

Regarding "Energy Demand," the report uses a baseline scenario and multiple sensitivity scenarios (take-off, efficient, and headwind scenarios) for modeling and forecasting. Data shows that in 2024, global data center electricity consumption accounted for about 1.5% of the global total (415 TWh). By 2030, it is projected to more than double, reaching approximately 945 TWh, accounting for less than one-tenth of global electricity demand growth, but its impact on local grids in specific regions (such as the US and China) is particularly significant. The report delves into the power supply mix needed to meet data center demand, pointing out that renewable energy and natural gas will play a leading role, with emerging technologies like small modular nuclear reactors and advanced geothermal also contributing. Meanwhile, factors such as grid congestion and tight equipment supply chains could put up to 20% of planned data center projects at risk of delay.

Regarding "Empowering Energy," the report systematically assesses the enormous potential of AI in optimizing existing energy systems and accelerating energy technology innovation. AI has already been applied to optimize oil and gas exploration and production, enhance grid balancing and renewable energy integration, improve industrial process efficiency, and refine traffic management and building energy consumption. In the field of innovation, AI is expected to significantly shorten the R&D cycles for new materials, battery chemistry, and carbon capture molecules. The report also analyzes the role and risks of AI in enhancing the resilience of energy systems (such as responding to cyberattacks and physical threats).

Finally, the report explores the broad impacts of AI development on energy security, investment patterns, the digital skills gap, challenges for emerging economies, and emissions. The report emphasizes that while AI's own energy consumption will lead to increased emissions, the emission reduction potential brought about by its widespread application could far exceed the former. However, this cannot replace proactive policy action. Faced with significant uncertainty, strengthening dialogue and cooperation between the technology sector and the energy industry is crucial. The International Energy Agency will continue to provide data and robust analysis to inform decision-making and help all parties better address the opportunities and challenges brought by the proliferation of AI.