Hyperscalers Face Natural Gas Price Risk as Forecast Warns of Tripling Costs
Original: Hyperscalers might regret embracing natural gas if new forecast proves correct
Why This Matters
AI infrastructure cost structures may shift dramatically if natural gas prices spike as forecast, affecting cloud pricing and energy markets broadly.
Energy research firm Noreva warns that natural gas prices could triple in parts of the U.S. as AI data center demand from Amazon, Google, Meta, and Microsoft collides with declining supply growth and rising LNG exports, potentially upending multi-gigawatt power plant investments.
Amazon, Google, Meta, and Microsoft have made major bets on natural gas to power their AI data centers, but a new forecast from energy research firm Noreva warns this strategy could prove costly. Meta announced a 7.5-gigawatt gas plant in Louisiana in March 2026, while Microsoft and Google each announced gigawatt-scale plants in Texas. Amazon followed with a 7.6-gigawatt Texas facility. Noreva CEO Peter Gardett warns that natural gas prices — currently ranging from $2 to $4.50 per million BTUs, with Henry Hub near $3 — could surge above $10 per million BTUs at certain delivery hubs. Gardett attributes the risk to three converging factors: hyperscaler AI demand, slower domestic supply growth as new wells become more expensive, and the integration of the U.S. domestic gas market with global LNG export markets. Since fuel accounts for roughly half the cost of electricity from large power plants, a doubling or tripling of gas prices could significantly raise AI inference costs or push hyperscalers back onto the grid, driving broader electricity prices higher. Gardett noted that at least one investor was 'surprised' by how much natural gas price risk hyperscalers are taking on, calling their behavior unusual for energy off-takers. Near-term futures contracts currently show no expectation of major price increases.