Gas Prices Could Triple, Upending Big Tech's Bet

Gas Prices Could Triple, Upending Big Tech's Bet

The News

The four biggest US cloud companies are building their own natural gas power plants to run AI data centers — and a new forecast says the fuel they are betting on could triple in price.

Energy research firm Noreva projects natural gas will climb above $10 per million BTUs at certain US hubs for extended stretches, according to a report given to TechCrunch and published August 14. Prices today run roughly $2 to $4.50 per million BTUs, with the widely traded Henry Hub in Louisiana just under $3.

It is a projection, not a finding. Futures markets currently price in no such spike.

"I think everyone in the energy markets has been lulled into a sense that gas prices can't go up," Noreva CEO Peter Gardett told TechCrunch. "You just need simple arithmetic to get to a much tighter gas market than you were in just a few years ago."

Why It Would Hurt

Fuel is about half the cost of electricity from a large power plant. So a doubling or tripling of gas prices makes "bring your own power" AI data centers far more expensive to run — which TechCrunch reports could push up the cost of AI tokens, or drive the hyperscalers back onto the public grid and push electricity prices higher for everyone else.

Gardett said the squeeze comes from three things stacking up at once: supply growth slowing as old wells fade and new wells get more expensive, US gas finally being plumbed into the global market through LNG exports, and what he called "the AI demand pull."

West Texas is the pinch point. Gas there has long been a cheap byproduct of oil drilling with nowhere to go. "They've finally built some pipelines out there, and a lot of that is headed towards export markets," Gardett said — connecting a discounted regional market to national and international prices, and creating the wide regional differentials he expects to drive some hubs past $10.

Timeline

March 2026

Meta says it will build a 7.5-gigawatt natural gas plant in Louisiana to power its Hyperion data center, per TechCrunch.

Days later

Microsoft and Google each announce their own gigawatt-scale gas plants, both in Texas.

August 8, 2026

TechCrunch reports Amazon plans a 7.6-gigawatt gas plant in Texas.

August 14, 2026

TechCrunch publishes the Noreva forecast exclusively.

Reactions

Gardett said at least one investor he spoke with was "surprised" by how much gas price risk the hyperscalers are taking on. "They're doing things that are not normal for an off-taker to do," he said. He also allowed that the bet is defensible on today's numbers: "It's not an unreasonable bet."

His forecast of how this ends up in shareholder letters: "On future Alphabet earning calls, you will hear them talk about the correlation between natural gas pricing and Google results, which is strange, but that's where we are."

Not everyone thinks the data center backlash is warranted. A column in The Blaze published August 15 argued America should take AI's risks seriously "without pretending every gallon, megawatt, or productivity gain is evidence of an approaching catastrophe."

What's Next

Watch the gas plants themselves — Meta's Louisiana build and the Microsoft, Google and Amazon projects in Texas are the exposure. Watch LNG export capacity and West Texas pipeline buildout, the two things Gardett says will link cheap regional gas to world prices. And watch quarterly earnings calls, where fuel costs would eventually show up.

Public patience is a separate clock. TechCrunch cites reporting that 80% of consumers are already worried about what data centers do to their utility bills — so far mostly about electricity.

More

The power math behind all of this: data centers could account for nearly 12% of US energy use by 2030, Gizmodo reported August 15, citing Lawrence Berkeley National Laboratory, with grid regulators warning demand growth is outrunning supply. Tech firms are chasing alternatives — researchers at Washington University published work August 6 in Nature Nanotechnology on a platinum-cobalt catalyst meant to make on-site hydrogen fuel cells durable enough for data center duty.

And the emissions ledger may be bigger than the buildings. Research published in npj Climate Action by former Microsoft sustainability staff Will and Holly Alpine, reported by Wired and Grist, models AI as a productivity booster for the oil and gas industry and finds it could raise global energy-related emissions by 1.2% to 4.8% — more than projected data center emissions themselves. Energy researcher Jon Koomey, who was not involved, called it "a credible attempt to answer that question using a macroeconomic model."

Colleges Condemn Students Mocking Kirk Assassination Anniversary

Charleston White Arrested Again on Florida Battery Charge

Border Patrol Rescues MS NOW Crew in Desert

Supreme Court Restores Missouri's Old House Map

Alaska Drops Felony Voter Charges Against Samoans

Trespasser Breaches Harris' Malibu Estate; No Arrest

Hegseth Pressed Downed Airmen Into 60 Minutes

Houthis Seize Mokha, Choking Saudi Oil Exports

Altman Rules Out OpenAI IPO This Year

Jury Acquits Lil Durk in Murder-for-Hire Case

Judge: Trump Broke Law Halving FEMA Staff

Maher Calls Democrats' Gaza Genocide Claim a Lie

Blizzard Revives StarCraft, Sends Diablo to Netflix

Trump's Queens Childhood Home Sells for $1.93M

Scientists Find New Virus in Tick Invading US

Britain Bans Settlement Imports; US States Threaten Payback

Prison Fire Kills 7 as Kurds Protest Killing

Kosovo Rallies for Ex-President Before Hague Verdict

Cruz Booed at GameDay Over College Sports Bill

Blast Hits Bulgarian Arms Plant Supplying Ukraine