The tech world has always had its share of FOMO, from dot-com mania to the blockchain boom. The latest frenzy? AI. Its hunger for power has sent companies racing to secure natural gas supplies and build massive plants to feed data centers.
Microsoft recently announced plans with Chevron and Engine No. 1 for a natural gas plant in West Texas that could eventually produce 5 gigawatts of electricity. Google is following suit with a 933 MW plant in North Texas, and Meta is adding seven more natural gas plants to its Louisiana Hyperion data center, bringing the site’s capacity to 7.46 GW — enough to power South Dakota.
The scramble is centered in the southern U.S., home to some of the world’s largest natural gas deposits. But the rush has created a shortage of turbines, essential for power plants. Wood Mackenzie predicts turbine costs could spike 195% by the end of this year compared to 2019. Delivery times are long, with new orders only possible by 2028.
Tech companies are betting that AI’s energy appetite will only grow, and natural gas will remain a reliable solution. But that confidence may be misplaced.
While U.S. natural gas supplies are abundant, production growth in major shale regions has slowed. Pricing details are largely undisclosed, leaving uncertainty about exposure to market swings. Electricity prices are tightly linked to natural gas, meaning large-scale AI data centers could inadvertently push prices higher for everyone.
It’s not just households at risk. Industries like petrochemicals, which rely heavily on natural gas and can’t easily switch to renewable sources, may find themselves competing with tech giants for resources. Extreme weather events could further strain supply, as seen in Texas in 2021 when freezing temperatures curtailed gas production.
By building their own plants, tech companies may appear self-sufficient, but in reality, they’re shifting the strain from the electrical grid to the natural gas network. Betting big on a finite resource could have unforeseen consequences, leaving these companies wondering if FOMO led them too far.
