Desperate for power, companies are pursuing new strategies to run data centers, from repurposing jet engines to beaming solar energy down from space.
AI companies pursue unconventional power—Crusoe's $1.25B jet turbines, Oracle's fuel cells, space-based solar—because utilities cannot meet demand fast enough. This energy bottleneck reveals AI infrastructure's critical constraints; on-site gas turbines already cause local pollution and raise consumer electricity costs.
AI firms are getting creative as they scramble for energy
Account
Donate
Clean energy journalism for a cooler tomorrow
Desperate for power, companies are pursuing new strategies to run data centers, from repurposing jet engines to beaming solar energy down from space.
As AI adoption accelerates, the key competitive bottleneck is shifting from models and GPUs to electricity itself. AI’s economics…
We’re announcing two partnerships to help deliver the reliable power our AI infrastructure and data centers need by advancing…
In the space between AI’s power and power demand lies opportunity.
A future where orbital data centers solve AI’s power crunch may be more distant than the hype suggests.
Emerald AI, which launched in July, is trying to solve the problem of powering the data centers that all AI companies need.
AI boom has energy suppliers scrambling to meet demand, but locating modular facilities close to wind and solar clusters could…
