AI is becoming an energy story as much as a software story. At the Energy and Innovation Summit in Pittsburgh on Tuesday, President Donald Trump and major energy and technology figures put electricity demand at the center of the next phase of artificial intelligence.
The message from the event was direct: more AI infrastructure means more power. For the fossil fuel industry, especially natural gas producers, that demand could create a new market at a time when parts of the gas business have been facing supply pressure and infrastructure challenges.
Trump links AI growth to a much larger power system
Trump acknowledged during his Pittsburgh speech that AI is “not my thing,” but said advisers had emphasized how important energy would be to the technology’s future. He recalled being told by “David”—most likely White House AI czar David Sacks—that the United States would need far more electricity to support AI.
“You need double the electric of what we have right now, and maybe even more than that,” Trump said. “I said, what, are you kidding? That's double the electric that we have. Take everything we have and double it.”
The summit brought together figures from both the technology and energy sectors. Panelists and attendees included David Sacks, Anthropic CEO Dario Amodei, Google president and chief investment officer Ruth Porat, and ExxonMobil CEO Darren Woods.
Companies announced $92 billion in investments across energy and AI-related ventures at the event. The timing fit a broader rush of investment around AI infrastructure. A day before the Pittsburgh meeting, Mark Zuckerberg shared on Threads that Meta would build “titan clusters” of data centers to support its AI work.
The cluster nearest to coming online, called Prometheus, is in Ohio and will be powered by onsite gas generation, according to a SemiAnalysis report cited in the source article.
Why Pennsylvania mattered
The summit’s location carried political and economic weight. Pennsylvania sits on the Marcellus and Utica shale formations, which helped drive the state’s fracking boom in the late 2000s and early 2010s. The state remains the country’s second-most prolific natural gas producer.
Natural gas had a visible role at the event. Toby Rice, CEO of Pittsburgh-based natural gas company EQT, moderated one of the panels and appeared onstage with Trump during the president’s speech. Rice calls himself the “people’s champion of natural gas.”
For the natural gas industry, AI data centers represent a potentially important source of new demand. The United States is the world’s top producer and exporter of liquefied natural gas, while global gas markets have been facing a mounting supply glut for years.
After a warm winter last year, Morgan Stanley predicted gas supply could reach “multi-decade highs” over the next few years. Large data centers could help absorb supply and support prices if their electricity demand becomes as large as industry advocates expect.
Data centers become a sales pitch for gas
Natural gas from Pennsylvania and the Appalachian region has faced two major headwinds described in the source: ultra-cheap natural gas from the Permian Basin in Texas and New Mexico, and limited infrastructure to move supply out of the region.
Clark Williams-Derry, an energy finance analyst at the Institute for Energy Economics and Financial Analysis, said those challenges explain why the industry is pushing the AI data center narrative.
These economic headwinds are “why the industry is doing their best to sort of create this drumbeat or this narrative around the need for AI data centers,” Williams-Derry said.
That argument is already shaping infrastructure plans. Pipeline companies are pitching new projects to truck gas from the northeast, saying they are responding to data center demand.
The Trump administration appears receptive to that framing. Since taking office, Trump has used AI as a reason to create new openings for fossil fuels, including a high-profile effort to resuscitate coal in the name of more computing power.
The summit reflected that priority. It was organized by Republican senator and former hedge fund CEO Dave McCormick, and no representatives from wind or solar companies appeared on any of the public panels.
Tech companies still want the cheapest power
The technology sector’s energy position is not identical to the administration’s. Tech companies have shown interest in any cheap power that can support AI, and they have quietly pushed back against some anti-renewables positions from the administration.
One announcement at the Pittsburgh summit underscored that split: Google announced a $3 billion investment in hydropower. The source article also notes that many tech giants have walked back climate commitments in recent years as AI has become a sharper focus, so the demand for cleaner energy is not necessarily driven by climate concern alone.
Economics are a major factor. Financial analyst Lazard said last month that utility-scale solar panels and batteries remain cheaper than building natural gas plants, even without tax incentives. Gas infrastructure also faces a global shortage, which changes the timeline for building new generation.
“The waiting list for a new turbine is five years,” Williams-Derry said. “If you want a new solar plant, you call China, you say, ‘I want more solar.’”
That tension appeared during one summit panel. Secretary of Energy Chris Wright, who previously led a fracking company, criticized the Obama and Biden administrations as being on an “energy crazy train” and dismissed their support for wind and solar. Speaking after Wright, BlackRock CEO Larry Fink said solar would likely support dispatchable gas in powering AI.
ExxonMobil CEO Darren Woods later gave some of the event’s limited attention to reducing emissions while discussing the company’s carbon capture and storage business.
The demand forecasts remain uncertain
Despite the political and industry momentum, the future scale of AI electricity use is not settled. Blackstone CEO Jonathan Gray said AI could help drive “40 or 50 percent more power usage over the next decade.” Porat cited projections from some economists that AI could add $4 trillion to the US economy by 2030.
Jonathan Koomey, a computing researcher and consultant who has contributed to research on AI and power, urged caution about projections like these.
“I view all of these projections with great skepticism,” Koomey said. “I don't think anyone has any idea, even a few years hence, how much electricity data centers are gonna use.”
In February, Koomey coauthored a report for the Bipartisan Policy Center warning that AI efficiency improvements and other technical developments make data center power load difficult to predict. He also said there are “a bunch of self-interested actors” involved in the AI power hype cycle, including energy executives, utilities, consultants and AI companies.
Koomey pointed to a historical parallel from the late 1990s, when investment banks, trade publications, and experts testifying in front of Congress warned that the internet could soon consume as much as half of US electricity. Some argued that more coal-fired power would be needed. A 1999 Forbes headline he cited read “Dig More Coal—The PCs Are Coming.”
That prediction did not come true. Efficiency gains reduced the internet’s energy needs, and Koomey said the early projections were based on faulty calculations.
The central question now is whether AI follows a different path or repeats the pattern: large claims, strong industry incentives, and uncertainty about what the power system will actually need.