Big tech needs to find more energy-efficient ways to run AI data centers, and direct their major energy investments, along with the government, to clean renewable energy that doesn’t make our ...

Within a few years, the most advanced AI models are projected to need five gigawatts of ... and whether a restart could ...

A few dozen ChatGPT queries cost a bottle’s worth of water. Tech firms should consider simpler solutions, like harvesting ...

In his 1982 book “Critical Path,” Buckminster Fuller calculated that humankind’s knowledge doubled every century up to 1900, ...

With the grid facing strain and more energy-hungry giants sitting down at the table, ensuring fair contribution isn’t just ...

Meta's plans for an AI data center running on nuclear power were complicated partly by the discovery of a rare bee species ...

AI requires power… a lot of it. Most AI queries are offloaded to massive data centers that consume exorbitant amounts of ...

Data centers use a lot of computational power, requiring a steady supply of water to cool the systems, which can be solved by ...

Environmental regulators reportedly quashed Mark Zuckerberg’s nuclear plant partnership meant to help power Meta’s ongoing artificial intelligence projects. Details remain scarce, but the main reason ...

Anna Erickson of Georgia Tech explains why the public and private sectors are investing in reactor technology and restarting retired plants. The system, though, still faces cost, safety and regulatory ...

Plans by Mark Zuckerberg’s Meta to build an AI data center in the US that runs on nuclear power were thwarted in part because ...

Big tech needs to find more energy-efficient ways to run AI data centers, and direct their major energy investments, along ...