Meta Platforms CEO Mark Zuckerberg has said that the company will invest hundreds of billions of dollars in constructing enormous artificial intelligence data centers for superintelligence, underlining his ambition to stay ahead in the highly competitive and rapidly growing AI space.

The first facility named Prometheus is expected to become operational in 2026, Zuckerberg said on the Threads platform as he outlined plans for multiple multi-gigawatt data centers. 

Its second installation called Hyperion will have the capacity to scale up to 5 gigawatts in the coming years, he added.

"We're building multiple more titan clusters as well. Just one of these covers a significant part of the footprint of Manhattan," the billionaire executive stated.

Zuckerberg has gone super aggressive in his efforts to stay ahead in the AI race since last month when Meta made a significant investment in data labeling company Scale AI.

As part of the deal, he also brought in Scale AI's Alexandr Wang to lead Meta's superintelligence unit as chief AI officer.

Earlier this month, Wang wrote on X that he will be working alongside former GitHub CEO Nat Friedman. Wang also listed the names of other employees hired by Meta from rival companies OpenAI and DeepMind. 

Later, Meta also roped in Safe Intelligence's co-founder and CEO Daniel Gross. This prompted Ilya Sutskever to  formally step into the role of CEO at Safe Intelligence, which he had launched last year after leaving OpenAI. 

Last week, Meta also roped in Apple's top executive Ruoming Pang in charge of artificial intelligence models.

Notably, last week, Meta also struck a deal to buy PlayAI, a voice technology startup, in its aggressive push into the AI space.

Meanwhile, competitors like OpenAI, DeepMind and xAI have been upgrading their own products, making acquisitions and hiring talent to strengthen their own AI game.