AI-Driven Data Center Expansion Accelerates Electricity Costs for U.S. Households

The data centers powering the rapid expansion of artificial intelligence are a significant factor behind rising household electricity bills, and relief may be some way off, energy specialists warn. Retail electricity prices for residential customers rose 7.4% in September to roughly $0.18 per kilowatt-hour, according to the most recent Energy Information Administration data. Forecasts from the EIA in May indicate that electricity costs are likely to outpace general inflation through at least 2026, with the impact falling unevenly across regions.

Analysts point to surging demand from electricity-intensive facilities that host cloud computing and AI workloads as a principal driver of the higher prices. These facilities are large installations of servers and supporting equipment that run generative AI, model training and inference tasks. The underlying dynamic is simple: demand — both current and projected — is growing faster than available supply and grid infrastructure.

Federal estimates underline the scale of the shift. The U.S. Department of Energy reported in December 2024 that data centers could consume between 6.7% and 12% of total U.S. electricity by 2028, up from 4.4% in 2023. John Quigley, a senior fellow at the Kleinman Center for Energy Policy at the University of Pennsylvania, described the trend as a “data center frenzy,” saying these facilities account for the bulk of recent increases in electricity demand and that the situation is likely to intensify.

Rising power costs have broader social and political consequences. Researchers at the Lawrence Berkeley National Laboratory noted that escalating electricity bills can strain household budgets, weaken economic competitiveness and complicate efforts to electrify sectors such as transportation and heating. Politicians across the spectrum are responding to affordability concerns: for example, New Jersey governor-elect Mikie Sherrill and Virginia governor-elect Abigail Spanberger both campaigned on pledges to lower electricity costs, with Spanberger warning against policies that could let data centers push up energy prices for other consumers. On the national stage, President Donald Trump has previously promised substantial reductions in energy costs.

Household financial stress is already measurable. A Century Foundation analysis found that the average overdue utility balance rose 32% from 2022, increasing from $597 to $789. Winter heating costs are also expected to climb: the National Energy Assistance Directors Association estimates that households relying on electric heating will face average winter bills near $1,205 this season, about 10% higher than the prior winter’s $1,093 projection. A Bank of America Institute report in October warned that consumers could feel renewed pressure on utility bills in the months ahead, particularly if the winter proves colder than average.

Global projections highlight the long-term implications. The International Energy Agency expects electricity demand tied to AI data centers to grow sharply, with some analyses suggesting a more than fourfold increase by 2030. Fatih Birol, the IEA’s executive director, has said that global electricity demand from data centers could more than double in the next five years and by 2030 equate to the present-day electricity consumption of an economy the size of Japan. The IEA also notes that the United States will see especially strong effects, with data centers accounting for a sizable portion of total demand growth.

Meeting that projected load will require substantial investment in new infrastructure — from transmission lines and substations to generation capacity — and the costs of those projects are often passed partially on to residential customers. Quigley observed that, in effect, households are subsidizing the buildout of AI-related compute capacity. Regional differences in grid management and exposure to extreme weather also shape local price outcomes; for instance, California has faced above-average price pressures tied to wildfire mitigation and insurance costs, the Lawrence Berkeley team found.

The trend interacts with other demand-side factors. Increased electrification, including the adoption of electric vehicles, is contributing to higher electricity consumption overall, a point underscored by the Bank of America Institute. Taken together, the rapid rise in AI-related power requirements and broader electrification trends create a complex planning challenge for utilities and policymakers seeking to balance reliability, affordability and decarbonization goals.