Over the last two years, America’s demand for electric power has surged thanks to the resurgence of U.S. manufacturing and the emergence of—and demand for—generative artificial intelligence (GenAI), which relies on power-hungry data centers. The emergence of these two substantial new electric loads is unprecedented, and policymakers must address this dual challenge rather than viewing industrial growth and data center growth as competing priorities that require choosing one over the other.
A promising solution is to transform data centers into grid and regional assets that offer flexible demand powered by carbon-free electricity. These centers can reduce energy consumption during grid strain, ramp up usage during renewable energy surpluses, and store excess electricity to feed back into the grid, improving its reliability and resilience.
To unleash this demand-side potential, the federal government, state policy makers, and utility regulators need to take three immediate actions: (1) fill data and knowledge gaps in AI data center design and operation; (2) improve AI data centers’ energy efficiency and integration with regional infrastructure and the grid; and (3) develop policies that transform demand-side strategies.
