Energy · resource ledger

Electricity

Data centers are a major new source of demand. Utility tariffs and cost allocation determine the household-bill effect.

AI-contributing11.8%LBNL reference-case share of U.S. electricity in 2030

What the evidence supports

LBNL’s 2026 update puts the 2030 data-center share at 11.8% in its reference case, with a 9.5%–15.3% scenario range. EIA says data-center growth is a major driver of rising U.S. load.

How the effect works: Large, flat loads require generation, transmission and distribution investment. Tariffs, utility regulation, fuel prices and dedicated-upgrade charges determine how those costs reach households.
Who pays or benefits: Data centers pay energy and connection charges directly. Regulatory cost allocation assigns shared system costs and fixed-cost revenue across customer classes.
What limits supply: The main challenge is how quickly new generation, transmission and substations can be built, connections can be studied and customer protections can be put in place.
Attribution boundary: National electricity prices do not show what bills would have been without AI. EIA found that natural-gas prices drove much of the 2025 increase in wholesale power prices.
Evidence that changes the grade: Any grade about household bills must be calculated utility by utility from approved rates and cost-allocation orders.

Sources

Public data, agency work and company reports

Claims related to this market

Companion references from the editor