natural gas is helping power the use of artificial intelligence

Most of us interface with artificial intelligence (AI) in countless ways every day. AI is not only key to many digital experiences (from text editing to online shopping to gaming), but also an important part of many industries, including health care and transportation.

As the use of AI grows, so does the need for energy to power data centers. These facilities are so energy intensive that, by one estimate, they could consume as much as 9% of all energy generated in the U.S. by 2030. Meeting this demand will require an abundant supply of reliable energy that is available 24/7.

In the U.S.—which supplies approximately one-quarter of the world’s natural gas—natural gas is helping to power that need.

In 2023, data centers consumed approximately 7.4 gigawatts (GW) of electricity globally, or enough to power more than six million homes.

Natural gas is one solution to powering these centers, and Chevron recently announced that it will work with Engine No. 1 and GE Verona to generate electricity for AI. The companies’ plans include building natural gas power plants directly connected to data centers.

During last year’s Gastech conference, Mike Wirth, Chevron chairman and CEO, highlighted the role that the Permian Basin could play in powering data centers.

“Natural gas will help power the rapid growth of artificial intelligence with its insatiable demand for reliable electricity,” Wirth said. “This means AI’s advance will depend not only on the design labs of Silicon Valley, but also on the gas fields of the Permian Basin.”

According to the U.S. Department of Energy (DOE), data centers are among the most energy-intensive building types. Right now, they account for approximately 2% of total electricity use in the U.S.—and that figure is expected to grow.

Data centers require so much energy because they involve running and cooling thousands of servers around the clock. These facilities consume 10 to 50 times the amount of energy per square foot of a typical commercial office building, DOE data indicates. Right now, these centers are pulling energy from the electrical grid, causing issues during peak demand and straining the system due to their high energy consumption.

In January, Microsoft said that it plans to spend $80 billion building AI-enabled data centers.

“More than half of this total investment will be in the U.S., reflecting our commitment to this country and our confidence in the American economy,” Brad Smith, Microsoft’s vice chair and president, said when announcing the investment.

By Blackstone’s estimates, more than $1 trillion will be invested in U.S. data centers through 2030. Amazon, Meta, and Google are among the tech behemoths building data centers throughout the U.S.

During an interview with CNBC, Wirth said the race for AI independence will require energy.

“The growth and demand for power can burden a grid that is already stressed. It could add cost to consumers,” he said. “Behind-the-meter solutions—powered by abundant American energy and natural gas—are an essential part of that equation going forward.”