Media365
Artificial intelligence is no longer just transforming technology; it is reshaping electricity markets and increasing power bills for millions of Americans.
The rapid expansion of AI data centers, packed with energy-intensive servers and graphics processing units (GPUs), has triggered an unprecedented surge in electricity demand, forcing utilities to invest heavily in new generation and transmission infrastructure. Experts say a significant share of those costs is ultimately being passed on to households through higher electricity rates.
According to the International Energy Agency (IEA), global electricity demand from AI-powered data centres is expected to more than quadruple by 2030. In the United States alone, data centres are projected to account for between 6.7 per cent and 12 per cent of the country’s total electricity consumption by 2028, up from 4.4 per cent in 2023, according to estimates by the US Department of Energy.
The surge in demand is already affecting consumers.
Residential electricity prices in the US rose 7.4 per cent year-on-year in September 2025 to around 18 cents per kilowatt-hour, according to the US Energy Information Administration (EIA). The agency expects electricity prices to continue rising faster than overall inflation through at least 2026.
AI Boom Driving Massive Power Demand
Modern AI models require enormous computing power for both training and inference. Even a single AI-generated video query can consume around 3.4 million joules of electricity—roughly equivalent to running a microwave oven for more than an hour.
To accommodate soaring demand, US utilities are planning an estimated $1.4 trillion in capital expenditure over the coming years to build new power plants and upgrade transmission networks.
Traditionally, these infrastructure costs are recovered through regulated electricity tariffs, meaning residential consumers often bear part of the financial burden even when demand is driven primarily by large commercial users.
In regions with significant data centre development, wholesale electricity prices have risen sharply, with some markets recording increases of up to 267 per cent.
Political Pressure Mounts
The rise in electricity costs has become a growing political issue.
State governments, consumer advocacy groups and attorneys general in states including Arizona, New York and Pennsylvania have questioned whether households should subsidise the rapid expansion of AI infrastructure.
Virginia, one of the largest data centre hubs in the world, has emerged at the centre of the debate. Politicians have argued that data centres should shoulder a greater share of grid expansion costs rather than passing them on to ordinary consumers.
The issue has also gained prominence ahead of elections as affordability remains a key concern for voters.
Tech Companies Respond
Facing mounting scrutiny, major technology companies including Google, Microsoft, Meta and Amazon have signed ratepayer protection commitments aimed at limiting the burden on households.
Under these pledges, companies have agreed to invest in new electricity generation and pay for infrastructure upgrades directly linked to their data centres, instead of relying solely on regulated utility charges.
AI Is Not the Only Factor
Energy analysts caution that AI is only one part of a broader challenge.
Electricity prices are also being driven higher by ageing grid infrastructure, the retirement of older coal-fired power plants, higher natural gas prices, extreme weather events and years of underinvestment in transmission networks.
However, analysts agree that AI has dramatically accelerated demand growth, exposing structural weaknesses in electricity systems that were already under strain.
As AI adoption continues to expand across industries, utilities, regulators and technology companies face increasing pressure to ensure that the costs of powering the digital economy do not fall disproportionately on households.
