EIA Projects Record U.S. Electricity Use Driven by AI Data Centers
Federal forecast expects commercial power consumption to surge as data center expansion strains electric grids.
The U.S. Energy Information Administration projected that nationwide electricity consumption will increase by 2% in 2026 and another 2% in 2027, reaching record highs. The federal forecast attributes the surge primarily to rapid expansion in artificial intelligence data centers and commercial facility power needs.
Key details
According to the EIA's latest Short-Term Energy Outlook, total U.S. electricity sales will reach 4,211 billion kilowatt-hours (kWh) by 2027. Demand growth is concentrated in the commercial sector, where electricity sales are expected to expand by 3.3% in 2026 and 2.7% in 2027. The commercial segment will account for 63% of total national electricity growth in 2026 and 56% in 2027.
To meet rising demand, utility-scale solar generation is projected to grow by 21% in 2026 and 18% in 2027, while natural gas generation increases by 2% in 2026 and 1% in 2027. Regional grid operators face distinct generation dynamics: demand in the PJM Interconnection accounts for nearly 45% of total U.S. generation growth, while solar expansion is concentrated in ERCOT (18 billion kWh added in 2026) and MISO (13 billion kWh added in 2026). Meanwhile, domestic power sector coal consumption is projected to fall by 8% in 2026 and 6% in 2027.
Why this matters
The forecast demonstrates that the rapid expansion of AI compute clusters is shifting national electricity consumption trends. By driving commercial electricity demand to historic highs, AI data centers require massive capital investments in generation capacity and transmission infrastructure, reshaping fuel mix projections and grid operations across major wholesale markets like PJM, ERCOT, and MISO.
Context
Historically, U.S. electricity demand grew at a modest pace of less than 0.5% annually. The sudden shift to 2% annual growth reflects the compounding power requirements of hyperscale data centers, high-density AI training clusters, and domestic manufacturing expansion. Regional grid operators are increasingly relying on natural gas and utility-scale solar buildouts to bridge immediate power supply gaps while mitigating domestic coal plant retirements.
What happens next
Federal regulators and regional transmission operators will monitor load interconnect queues to manage reliability risks. As commercial electricity sales rise toward 4,211 billion kWh, state public utility commissions are expected to evaluate new large-load tariffs and rate structures to shield residential consumers from bearing the costs of AI-driven power infrastructure upgrades.
Source: Utility Dive Published on AI Usage Global, author: AUG Bot
