Trump and the energy industry are eager to use fossil fuels to power AI

The industry is finding a willing partner in the Trump administration. Since taking office, Trump has used AI as leverage to open up opportunities for fossil fuels, including a meticulous effort to restore coal in the name of more computing power. The summit organized by Republican Senator (and former hedge fund CEO) Dave McCormick clearly reflects the government’s priorities in this regard: There is no representative of any wind or solar company on any public panel.
Tech companies have expressed interest in using any and all the cheap electricity available to AI and quietly oppose anti-renewable energy positions in certain administrations, not necessarily on the same page as the Trump administration. In the announcement released at the summit, there was $3 billion in Google’s investment.
This demand is not necessarily driven by a significant focus on climate – with the focus on AI in recent years, many people’s climate commitments have shrunk their climate commitments – but rather purely economics. Financial analyst Lazard said last month that installing utility-scale solar panels and batteries is still cheaper than building natural gas power plants, even without tax incentives. Natural gas infrastructure is also facing global shortages, which makes the time to set up power generation very different.
“The waiting list for new turbines is five years,” Williams-Derry said. “If you want a new solar power plant, you call China, you say, ‘I want more solar.’”
Given the ideological divisions at the summit, the situation occasionally becomes a bit awkward. In a group, Energy Secretary Chris Wright led a fracking company before coming to the federal government, where he discussed in detail how the Obama and Biden administrations took the “energy crazy train” to ridicule with support from these administrations for wind and solar. BlackRock CEO Larry Fink spoke directly after Wright, acknowledging that solar might support dispatchable gases when powering AI. Incredibly, Woods, a panelist at Exxonmobil CEO, later paid some of the only verbal services to reduce the idea of emissions heard throughout the event. (Woods is touting the oil giant’s carbon capture and storage business.)
Still, most of the time, the hype train went smoothly for the most part, and everyone agreed with one thing: we need a lot of power soon. Blackstone CEO Jonathan Gray said AI could help “use 40% or 50% of power usage over the next decade,” while Google’s Porat mentioned some economists’ forecasts that AI could add $4 trillion to the U.S. economy by 2030.
It’s easy to find a wide variety of headlines or reports (usually based on forecasts produced by private companies) for massive growth figures in AI. “I look at all these predictions with great skepticism,” said Jonathan Koomey, a computer researcher and consultant. “I don’t think anyone has any idea, even a few years, so how much power data centers are to be used.”