energy ai tech

Tech Giants' Natural Gas Gamble Could Cost Billions

Amazon, Google, Meta, and Microsoft are betting on natural gas to power AI data centers, but prices could triple as demand surges and supply tightens.

Tech Giants' Natural Gas Gamble Could Cost Billions

After years of chasing renewable energy deals, the world’s biggest tech companies are making a dramatic pivot. Amazon, Google, Meta, and Microsoft have collectively announced plans to build massive natural gas power plants to fuel their AI ambitions. But a new research report suggests this could be one of the costliest mistakes in tech history.

Energy research firm Noreva warns that natural gas prices could triple in some U.S. regions over the coming years. Today, prices hover around $2 to $4.50 per million BTUs at major hubs. Noreva expects them to soar above $10 in certain markets. For hyperscalers who’ve locked in massive quantities at today’s prices, that’s a nightmare scenario.

“I think everyone in the energy markets has been lulled into a sense that gas prices can’t go up,” Peter Gardett, CEO of Noreva, told TechCrunch. Simple math tells a different story. Hyperscalers are consuming unprecedented amounts of natural gas just as supply growth is slowing and liquefied natural gas exports are accelerating.

The Cheap Gas Trap

Why did these companies make such aggressive commitments to natural gas? Blame rock-bottom prices. Meta announced a 7.5-gigawatt gas power plant in Louisiana. Microsoft and Google each revealed gigawatt-scale plants in Texas. Amazon countered with a 7.6-gigawatt facility, also in Texas. These aren’t small bets. For companies historically allergic to massive capital expenditures, this represents a seismic shift.

West Texas made these deals irresistible. Oil wells there produce natural gas as a byproduct, and for years it went cheap because there was nowhere to sell it. New pipelines changed that equation, suddenly connecting regional producers to national and global markets. Hyperscalers rushed in, thinking they’d found free money.

“They’re doing things that are not normal for an off-taker to do,” Gardett said, noting that even seasoned energy investors were surprised by the risks these tech companies were willing to stomach.

When Prices Explode

Fuel costs roughly half the price of electricity from a large power plant. Double or triple natural gas prices and suddenly those shiny new AI data centers become expensive to operate. Hyperscalers will face an ugly choice: dramatically raise token prices to cover energy costs, or abandon their “bring your own power” strategy and connect to the grid. The latter option could spike electricity prices for everyone else, reigniting public backlash.

Already, 80% of consumers worry about data centers’ impact on their utility bills. Most concerns center on electricity consumption. Adding natural gas price volatility to that equation could trigger a PR disaster for companies already facing regulatory scrutiny.

The Global Gas Market Meets AI

What makes Gardett’s warnings credible is his explanation of why prices will rise. For decades, natural gas supply remained relatively stable because demand was flat. New wells replacing declining production kept the market balanced. Those dynamics are breaking down on two fronts.

First, new wells are getting more expensive to develop. Energy companies can add more supplies, but not at the breakneck pace of the past. Second, and more importantly, the domestic gas market is finally connecting to the global market through LNG exports and new pipelines.

This matters enormously. When West Texas becomes linked to international prices, regional dynamics change everywhere. “You will get places where you get a lot of gas next to someplace where there’s none, and so you’ll get those big differentials,” Gardett explained. Those price differentials will spike in some regions, pushing costs above $10 per million BTUs.

Futures markets aren’t yet pricing in this risk. Most contracts assume prices remain stable. “It’s not an unreasonable bet,” Gardett acknowledged. But the CEO clearly disagrees with the market’s complacency.

Hyperscalers have ventured into territory they barely understand. Energy markets operate under different rules, with different timescales and different risks. On future earnings calls, executives will have to explain why natural gas prices matter to investors. That conversation should be happening right now.

Source: TechCrunch

Filed under
energyaitech