AI data center 'frenzy' is pushing up your electric bill — here's why (www.cnbc.com)

Rapid expansion of AI data centers is materially increasing U.S. electricity demand and contributing to higher residential prices: retail electricity was up 7.4% in September to roughly $0.18/kWh, and the EIA expects electricity prices to outpace inflation through at least 2026. Energy experts point to a “data center frenzy” as a primary driver—DOE estimated data centers will consume 6.7–12% of U.S. electricity by 2028 (up from 4.4% in 2023), while the IEA projects global AI data-center demand could more than quadruple by 2030, equating to as much electricity as today’s Japan. Forecasted demand is already prompting costly grid upgrades (power lines, substations, generation) that utilities and developers partly pass on to households. The significance for AI/ML is twofold: large-scale model training and inference are changing the economics and physical constraints of compute deployment, forcing operators to factor in grid capacity, regional energy prices, and policy scrutiny. Impacts are uneven—states in the West and Northeast face sharper price hikes due to regional grid stress, wildfire mitigation costs, and extreme weather—while rising bills are straining household budgets (average overdue utility balances rose ~32% since 2022). The trend amplifies policy debates about siting, fair cost allocation, electrification goals, and whether data-center growth should be subsidized by residential customers or guided by demand-response, efficiency, and targeted infrastructure investment.