AI companies are building giant natural gas plants to power their data centers. What could be wrong?


Who doesn’t love a good FOMO tour? From dot-com to Web 2.0, virtual reality to blockchain, the tech industry has had its share of being too afraid to miss a trend.

The AI ​​bubble is the big daddy of them all. His first generation – in a hurry lock down power for data centers – now causing a mad rush to secure natural gas supplies and equipment. If FOMOs could have babies, then the AI ​​bubble already has grandchildren.

Microsoft said Tuesday that it is working with Chevron and the No. 1 engine build a natural gas power plant It could generate 5 gigawatts of electricity in West Texas. Google this week confirmed Working with Crusoe to build a 933 MW natural gas power plant in North Texas. Last week, Meta announced that it had added seven more natural gas power plants to its Hyperion data center in Louisiana, bringing the site to 7.46 GW. enough to power the entire state of South Dakota.

Missing someone?

Recent investments have been concentrated in the southern United States, home to the world’s largest natural gas deposits. The US Geological Survey recently estimated that there is enough energy in one region to power the entire United States. 10 months with himself. Every data center operator wants a piece of that.

The panic for natural gas has led to a shortage of turbines for power plants, and prices are expected to increase by 195% from 2019 prices by the end of this year. according to Wood to Mackenzie. Equipment accounts for 20% to 30% of the cost of a power plant. Companies won’t be able to place new orders until 2028, and turbines take six years to deliver, the consultant notes.

This means that tech companies are betting that the AI ​​fever will not go away, that the need for AI will continue the amount of exponential powerand natural gas production will be essential to success in the age of artificial intelligence.

Techcrunch event

San Francisco, CA
|
October 13-15, 2026

They may regret the third assumption.

Although natural gas is abundant in the US and the fuel is not cheap to transport, the country remains somewhat insulated from the turmoil in the Middle East. But supplies are not unlimited, and three major regions responsible for three-quarters of all U.S. shale gas production have recently seen production increases. slowed down considerably.

It’s unclear how the tech companies are insulated from the price swings, as none of them have disclosed the specific terms of their contracts. Much will depend on how firm the price is in these contracts.

Even if negotiated prices are as firm as possible, companies can still face backlash.

Because natural gas produces about 40% of electricity in the United States. according to According to the Energy Information Administration, electricity prices are closely related to natural gas prices. Tech companies can shield themselves from some scrutiny by moving gas power plants behind the meter—bypassing the grid and connecting them directly to data centers. But natural gas is not an unlimited resource, and if their ambitions get too big, even behind-the-meter operations can raise energy prices for everyone. We’ve all seen how this plays out.

This will not only concern ordinary households. Other industries, including those that depend more on natural gas and cannot yet turn to renewable energy sources, may shy away from resource-intensive data centers. It’s easy to power a data center with wind, solar, and batteries. Do you run a petrochemical plant? Not so much.

Then there’s the weather. A cold winter can change the calculation by increasing demand among households. Wellheads can freeze, severely depressing reserves, As it happened in Texas in 2021. When gas runs out, suppliers will be faced with a choice: keep AI data centers running or keep people heating their homes?

By cutting off natural gas supplies and going behind the meter, tech companies can claim they’re “bringing their own power” without straining the power grid. But in reality, they simply transfer their usage from one network to another, to the natural gas network. The AI ​​trap demonstrated just how physically limited the digital world remains. Does it make sense for them to bet big on a limited resource? Tech companies may regret falling into FOMO.



Source link

Leave a Reply

Your email address will not be published. Required fields are marked *