Meta invest billions in data centers to boost AI development

By Alimat Aliyeva

Meta founder Mark Zuckerberg has announced plans to invest hundreds of billions of dollars in building massive artificial intelligence data centers across the United States, Azernews reports.

According to Zuckerberg, the first multi-gigawatt facility, named Prometheus, is expected to go online in 2026. One of the planned sites will span an area of 59.1 square kilometers — nearly the size of Manhattan.

Meta is making a major bet on the development of so-called “superintelligence” — a form of AI that, in the company’s view, could eventually surpass even the most intelligent humans. The company, which primarily generates revenue through online advertising, brought in over $160 billion last year.

In a post on Threads, Zuckerberg revealed that Meta is building several AI clusters, each capable of handling immense workloads. One of these, called Hyperion, is projected to scale up to five gigawatts in the coming years — making it one of the most powerful computing hubs in the world.

Support these ambitious AI goals, Meta is also reportedly working on designing its own custom AI chips and networking infrastructure — a move aimed at reducing reliance on external hardware providers and boosting efficiency at scale.