(Jack_Ty_Spro/Shutter Stock)
The rise of AI is putting real pressure on the US power grid. As the demand for computations, the data centers behind the AI system are becoming the country’s top energy users. They are now essential for digital life, but with the use of their non -stop energy, it is starting to reorganize how the cost of electricity is distributed across the country.
An investigation by Bloomberg has revealed that wholesale electricity prices have increased dramatically in areas near large data centers. In one example, a month’s prices were 267 % higher than five years ago. In the highest addictive sites, about 75 % of the data centers were located 50 miles from the dense cluster. This is spread through regional grid, which means that households and businesses are paying high costs.
The data behind each AI application is a constant stream. These centers do not just take action on information – they absorb, store, store and move in large quantities. GPUs operates non -stop training and analogy jobs, while the Peta Byte Scale Storage Models holds datasters. This activity produces severe heat, and requires a cooling system to manage it, which uses energy use even more. The model is always made for development and is made for development, which transforms each facility into a permanent electric drain on the grid.
Bloomberg analysis points to a massive change. By 2035, AI -backed data centers can use close to 9 % of all US electricity. The number looked ridiculous a few years ago. But it’s not just about how many machines are online. The important thing is that the data movement, the need for speed, and the great infrastructure behind it. This load is sending prices upward, combined with the elderly grid system, even in places where there is no AI center.
(Phil Vector/Shutter Stock)
Most of the recent growth is focused on places like Iowa, Georgia and parts of the southwest. These regions offer affordable land, low obstacles, and relatively easy access to electricity. However, many of these people were never designed for the use of such industrial electricity. Even in areas with no facilities, there is a tension now, as the strength of the neighboring grid is diverted to fulfill the burden.
According to Mark Christie, a former chair of the Federal Energy Regulatory Commission, he clearly said, “The reliable crisis is now here. It is not at a distance.” It is referring to blackout and grid instability, which can happen when energy demand increases rapidly when utility can create a infrastructure to help it. In many cases, the utility is being flooded by the new data centers with coordination requests, which are increasing the speed of speeding speeds of planning cycle.
This pressure is not unique to the United States in Japan, the power auction prices climbed to record heights as the government is ready to accelerate the burden of AI work. Malaysia has recently increased data center power rates due to local shortage. In the UK, Aurora Energy Research estimates that the demand for electricity from AI infrastructure could increase electricity prices by 9 % by 2040.
According to Bloombergenf, worldwide, data centers can use more than 4 % of all electricity by 2035. If these facilities were a country, they would be ranked fourth in energy consumption – only behind China, the United States and India. In the United States alone, it is expected that by 2035, the use of AI -related data center will increase from 4.5 4.5 percent to about 99 percent. Some experts have been calling it the biggest jump in the demand for electricity since the arrival of air conditioning in the 1960s.
(Best web/shutter stock)
David Crane, CEO of Generat Capital and former Advisor to Biden Energy, said: “Without a reduction, data centers are making things really expensive for the rest of Americans to suck all the load.” He warned that some US areas may face brownout early next year if the demand is increasing without any integrated grid response.
Meanwhile, politics is forming on how the country reacts. Some leaders have promised to reduce electricity prices by 50 %, but domestic bills have only increased. Federal policies have recently received support for solar and storage. Bloombergenf is now expected to reduce the annual rollout of clean energy projects by 2035 to 23 %, which is under earlier goals.
At the same time, the new investment is being presented in coal, gas and nuclear sources, and the moving path is complicated. AI infrastructure is growing rapidly and the grid is already under pressure, the cost of the data is no longer measured in the dollar. It is also being measured in Klawat.
This article first appeared on our sister’s publication, Big Datawire.
Relevant
