This report comes following a recent meeting OpenAI Chief Executive Officer Sam Altman had at the White House. OpenAI's purpose at this meeting was to put forward to the administration of President Joe Biden and Vice President Kamala Harris the economic and national security benefits of building dozens of five-gigawatt data centers in various U.S. states. (Related: Microsoft to bring infamous nuclear plant back to life to feed POWER-HUNGRY AI data centers.)
To put in context the energy needs of OpenAI's proposal, just one data center that uses up five gigawatts of electricity each year would need the annual energy output of around five nuclear reactors, or enough energy to power almost three million homes.
According to OpenAI, investing in the construction of these data centers could provide tens of thousands of new jobs to Americans, boost the U.S.'s gross domestic product and ensure that America maintains its lead in the development of AI technology. To achieve this goal, OpenAI is demanding government support for policies that promote greater data center capacity.
Altman has also been lobbying investors to help fund the very costly infrastructure needed to support his company's growth and the growth of AI technology.
"OpenAI is actively working to strengthen AI infrastructure in the U.S., which we believe is critical to keeping America at the forefront of global innovation, boosting deindustrialization across the country and making AI's benefits accessible to everyone," OpenAI said in a statement.
OpenAI having trouble finding enough power for just one data center
Joseph Dominguez, president and CEO of Baltimore, Maryland-based electricity provider Constellation Energy, told Bloomberg News that Altman is proposing the construction of five to seven data centers, each requiring around five gigawatts of power annually. This number has not been confirmed by OpenAI, and the document the company shared with the White House regarding its plan does not provide a specific number.
Dominguez said OpenAI's first goal is to focus on the construction of a single data center to show that building it would be beneficial to the United States. Afterward, Altman plans to expand from there based on the resources available.
"Whatever we're talking about is not only something that's never been done, but I don't believe it's feasible as an engineer, as somebody who grew up in this," said Dominguez. "It's certainly not possible under a timeframe that's going to address national security and timing."
OpenAI is seeking help from the government and investors regarding the power needs of its data centers at a time when new energy projects in the U.S. are facing significant delays due to a variety of factors, including permitting delays, supply chain issues, labor shortages and the very long wait times needed before these new projects can be connected to the U.S. energy grid. Energy executives like Dominguez have further stressed that powering just one five-gigawatt data center would be a major challenge.
John W. Ketchum, chairman, president and CEO of renewable energy giant NextEra Energy, said, without naming any specific companies, that he had received requests from some major Big Tech companies to find sites that can support five gigawatts of demand.
"That's the size of powering the city of Miami," Ketchum remarked. He noted that it would be easier to find a place in the U.S. that can fit a one-gigawatt data center, and finding five would require a mix of new wind and solar farms, battery storage and a connection to the wider grid.
Watch this clip from CNBC discussing how the construction of new AI data centers all over the world is fueling a boom in cooling technology to help prevent these data centers from overheating.
This video is from the TrendingNews channel on Brighteon.com.
More related stories:
Big Tech eyes nuclear power to meet the demands of AI computing.
Big Tech has a growing appetite for America's electricity and water resources.
America at risk of running out of electricity due to Big Tech's growing demand for power.
Microsoft acquires over 1,000 acres of land in Wisconsin for data center campus.
Sources include:
Please contact us for more information.
