In the United States, amid rapid development of AI technologies there is an increase in electricity consumption and, as a result, an increase in electricity prices.
It is noted, that the advice and assistance provided by AI agents, which are offered to users in large numbers by leading american technology companies, requires paying large electricity bills to support the operation of data centers. At present, electricity prices for Americans continue to rise, as the country’s energy system was not ready for such a surge in demand.
According to Newsweek, rising data center consumption has already led to a 6.5% increase in electricity prices between May 2024 and May 2025. Average growth rates in states, such as Connecticut and Maine, made up 18.4% and 36.3%, respectively. Growth is expected to continue as tech companies continue to expand the infrastructure needed to run AI systems.
“To keep up, utilities are increasingly using outdated fossil fuel power plants to generate enough electricity to meet growing demand. Dominion Energy, which serves most of Virginia, has asked regulators to require high-load consumers to pay a fairer share of the cost of grid modernization. Without reforms, electricity prices in some parts of Virginia are expected to increase by 25% by 2030″, — Newsweek journalists note.
Hardly the problem can be solved by developing more efficient AI models. The fact is that the AI tools available today are based on the constant acquisition of information, which forces the companies that created them to extract as much content as possible. Continuous data collection and improvement of AI models require more and more energy.
By report The Lawrence Berkeley National Laboratory’s “U.S. Data Center Energy Consumption 2024”, the average annual growth rate of electricity consumption from 2014 to 2018 was 7%. From 2018 to 2023, the growth rate was already 18%, and from 2023 to 2028, this figure is expected to be between 13% and 27%.
This will result, in AI data centers and other infrastructure consuming between 6.7 and 12% of the total energy consumption in the United States. American power grids are still not ready to meet this demand.
How note Reuters journalists, the rapid expansion of data centers processing large amounts of information for artificial intelligence and cryptocurrency mining is forcing power grid operators to plan for contingencies and complicating the already difficult task of balancing supply and demand in the market. The report mentions several emergency situations caused by the inclusion of own generators in data centers.
The operation of these generators can lead to an excess of electricity and grid overload if the companies operating the relevant infrastructure do not respond quickly enough. Failure to respond quickly could lead to cascading blackouts across the region.
Even living near data centers that consume large amounts of electricity shortens the life of electrical appliances and can lead to problems in their operation, overheating, and fire. All this demonstrates that the current obsession with artificial intelligence creates huge demands on power grids. This leads to an additional load on the devices connected to these networks and increases the risk of power outages due to the use of on-site generators designed solely to protect the facility’s equipment without taking into account, the impact it can have on everyone else.
Source: Tom’sHardware
Spelling error report
The following text will be sent to our editors: