AI Needs More Abundant Power Supplies to Keep Driving Economic Growth
IMF Blog, May 13, 2025
Source details
- Canonical URL
- AI Needs More Abundant Power Supplies to Keep Driving Economic Growth
Other formats
Bibliographic details
- Authors: Christian Bogmans, Patricia Gomez-Gonzalez, Giovanni Melina, Sneha Thube
- Published: May 13, 2025
Summary
- Artificial intelligence is an emerging source of productivity and economic growth and is reshaping employment and investment.
- AI requires growing amounts of electricity for data centers, creating major implications for global electricity demand and power grids.
- The blog summarizes analysis included in the IMF’s April 2025 World Economic Outlook and an IMF working paper: Power Hungry: How AI Will Drive Energy Demand. For broader AI impacts, it cites the working paper: The Global Impact of AI: Mind the Gap.
Key statistics and projections
- World data centers consumed as much as 500 terawatt-hours of electricity in 2023 (Organization of the Petroleum Exporting Countries estimate).
- The 2023 total was more than double the annual levels from 2015-19.
- OPEC projects data center electricity consumption could triple to 1,500 terawatt-hours by 2030.
- By 2030, electricity used by data centers alone would be comparable to that of India, the world's third-largest electricity user.
- Data center consumption by 2030 would be 1.5 times as much power as projected consumption by electric vehicles.
- Power needed for US server farms is likely to more than triple, exceeding 600 terawatt-hours by 2030 (medium-demand scenario projection by McKinsey & Co).
- Under current energy policies, the AI-driven rise in electricity demand could add 1.7 gigatons in global greenhouse gas emissions between 2025 and 2030 — about as much as Italy’s energy-related emissions over a five-year period.
Main implications
- Rapid growth in electricity demand from AI can strain power grids and require large additions to supply capacity.
- If electricity supply is responsive, increased demand will stimulate supply and lead to only a small increase in power prices.
- If supply responses are sluggish, power prices could rise steeply, hurting consumers and businesses and possibly curbing growth of the AI industry itself.
- The boom in building new data center facilities underscores urgency for effective energy strategies to ensure adequate supplies for surging demand.
Uncertainties and scenarios
- Demand for computing and electricity from AI platforms is subject to wide uncertainty.
- Efficient, open-source AI models like DeepSeek lower computing costs and electricity demand.
- Lower costs may increase AI usage, offsetting efficiency gains.
- More energy-intensive reasoning models raise electricity demand.
- Net effect on electricity demand is uncertain; this uncertainty may delay energy investments and cause higher prices.
Policy recommendations
- Policymakers and businesses must collaborate to ensure AI achieves its full potential while minimizing costs.
- Implement policies that incentivize multiple energy sources to:
- Enhance electricity supply.
- Help mitigate price surges.
- Contain emissions.
Source: AI Needs More Abundant Power Supplies to Keep Driving Economic Growth, IMF blog, May 13, 2025.
Content in this bundle
- Commodity Special Feature