Microsoft, Google, and Meta Bet on Natural Gas: A Risky Gamble?
Le brief IA que les pros lisent chaque soir
Les 7 actus IA du jour, décryptées en 5 min. Gratuit.
Inclus dès l'inscription : notre sélection des meilleurs guides & comparatifs IA.
Choisis ton rythme
Gratuit · Pas de spam · Désabonnement en 1 clic
A Rush for Natural Gas to Power AI
In the tech world, the fear of missing out, often referred to by the acronym FOMO, has marked several periods in history, from the dot-com bubble to the rise of blockchain. Today, this fear manifests as a rush for natural gas to power the data centers necessary for artificial intelligence. Microsoft, Google, and Meta are at the forefront of this trend, seeking to secure substantial energy supplies.
Microsoft recently announced a partnership with Chevron and Engine No. 1 to build a natural gas plant in West Texas, capable of producing up to 5 gigawatts of electricity. Meanwhile, Google confirmed its collaboration with Crusoe to establish a 933-megawatt plant in North Texas. Meta, for its part, has decided to add seven natural gas plants to its Hyperion data center in Louisiana, bringing the total capacity of the site to 7.46 gigawatts, enough to power the entire state of South Dakota.
Concentrated Investments in the Southern United States
These massive investments are primarily concentrated in the southern United States, a region rich in natural gas reserves. The U.S. Geological Survey recently estimated that the gas reserves in this area could supply the entire United States for ten months. However, this rush for natural gas has led to a shortage of turbines for the plants, with prices expected to rise by 195% by the end of the year compared to 2019, according to the firm Wood Mackenzie. The necessary equipment accounts for 20% to 30% of the total cost of a plant.
The demand for turbines is such that companies will not be able to place new orders until 2028, and it will take six years to receive this equipment. This illustrates how tech companies are betting on a growing demand for energy for AI, but this bet carries risks.
The Stakes and Risks of Energy Dependence
Although the United States has abundant natural gas resources, production in the three major regions responsible for three-quarters of all shale gas production in the U.S. has significantly slowed. Fluctuations in natural gas prices could also affect electricity costs, which depend on this resource for 40%.
By disconnecting from the traditional power grid to connect directly to their own plants, these companies hope to avoid price hikes. However, this strategy could have repercussions on electricity prices for consumers and other industries, particularly those that cannot yet turn to renewable energy sources.
Climate and Economic Challenges
Weather conditions, such as a harsh winter, could also disrupt the supply of natural gas, as occurred in Texas in 2021. In such a scenario, suppliers will have to choose between powering AI data centers or heating homes. When gas runs short, companies may face difficult choices, with implications for the general population.
Ultimately, by focusing on natural gas, tech companies are merely shifting their energy dependence from one grid to another. This strategy could prove problematic in the long term, as the digital world continues to rely on limited physical resources. Tech companies may regret succumbing to the pressure of FOMO, betting on a resource that is not infinite. The consequences of these choices could be felt far beyond the tech sector, significantly impacting the economy and the environment.
Brief IA — L'actualité IA en français
L'essentiel de l'actualité de l'intelligence artificielle, décrypté et expliqué chaque jour.