Signal

Data centers are turning to old jet engines as the wait for power grows

First reported by TechCrunch ·

The signal ●●○○ Compiled by AI from TechCrunch, Tom's Hardware, The Daily Caller and Techstrong IT
Why you might care

New data center power generation options are now available in under two years.

What happened

Data centers are increasingly turning to repurposed jet engines to generate their own electricity due to significant delays in securing grid connections. These converted GE engines, repurposed by ProEnergy, can power a data center within approximately 24 months, a stark contrast to the seven years it can take to connect to the grid in areas like Northern Virginia. Companies such as Elon Musk's xAI and OpenAI are already deploying these units, with xAI using eight in Mississippi for its Memphis data center and OpenAI planning for its Stargate site in Abilene, Texas. Startups like Crusoe are also investing heavily in this technology, though trade-offs include lower efficiency and higher pollution compared to conventional power plants.

What it means

The reliance on converted jet engines highlights a critical bottleneck in the AI infrastructure race: power availability. As demand for AI compute surges, the traditional utility infrastructure is struggling to keep pace, forcing companies to seek rapid, albeit less conventional, power solutions. This trend suggests a short-term workaround that prioritizes speed-to-market over long-term efficiency and environmental sustainability, potentially shaping the next phase of data center development.

This situation creates an immediate competitive advantage for AI companies that can secure these alternative power sources, enabling them to launch and scale operations faster than those dependent on lengthy grid connection timelines. However, the long-term viability of jet engines as a primary data center power solution remains questionable due to their inefficiency and environmental impact, suggesting a potential shift back to grid power or more advanced, sustainable solutions as they become available.

AI-written summary. May contain errors.

Data