Zuckerberg Commits Hundreds of Billions to AI Data Centers in Bold Superintelligence Bid

In a sweeping declaration of Meta Platforms’ ambitions in artificial intelligence, CEO Mark Zuckerberg announced plans to invest hundreds of billions of dollars to construct vast AI data centers dedicated to the development of superintelligence.

The social media and technology giant is pushing forward with a scale and urgency that underlines its intention to dominate the next era of computing.

Zuckerberg revealed that Meta’s first multi-gigawatt data center, named Prometheus, is on track to begin operations in 2026. A second data center, called Hyperion, will follow and is designed to eventually scale up to 5 gigawatts in capacity. “We’re building multiple more titan clusters as well. Just one of these covers a significant part of the footprint of Manhattan,” Zuckerberg wrote in a post on his Threads platform.

Meta is among the major tech firms racing to develop advanced AI models capable of surpassing human cognitive performance across various tasks—a concept broadly referred to as superintelligence. As part of this effort, the company has engaged in a fierce talent war, securing multi-million-dollar contracts for top engineers and striking high-profile industry deals.

Zuckerberg also cited a report from SemiAnalysis that suggested Meta could be the first AI lab to bring a gigawatt-plus supercluster online—an engineering milestone in the race for AI supremacy.

Despite mounting investor questions about whether such massive capital outlays will pay off, Zuckerberg remained confident, citing Meta’s robust advertising business as the foundation for the expansion. “We have the capital from our business to do this,” he stated.

The announcement comes as Meta stock traded 1 percent higher, with the shares having gained over 20 percent since the start of the year. The company posted $165 billion in revenue in the previous fiscal year, reinforcing its ability to self-fund ambitious initiatives.

To better align its AI ambitions, Meta recently reorganized its AI operations under a new division dubbed Superintelligence Labs. This shift followed internal struggles with the open-source Llama 4 model and the departure of key personnel. The new unit is expected to focus on developing revenue-generating products such as the Meta AI app, image-to-video ad tools, and smart glasses.

There is also speculation around a strategic pivot in Meta’s AI development approach. According to a separate report from The New York Times, senior members of Superintelligence Labs have been debating whether to abandon Behemoth, the company’s most powerful open-source model, in favor of a closed alternative.

D.A. Davidson analyst Gil Luria acknowledged Meta’s heavy investment but noted the long road ahead: “Meta was investing aggressively in AI as the technology has already boosted its ad business by allowing it to sell more ads and at higher prices. But at this scale, the investment is more oriented to the long-term competition to have the leading AI model, which could take time to materialize,” Luria said.

In a further display of commitment, Zuckerberg has personally spearheaded a high-stakes talent acquisition campaign to build the team at Superintelligence Labs. The group will be led by former Scale AI CEO Alexandr Wang and former GitHub CEO Nat Friedman, following Meta’s $14.3 billion investment in Scale.

As part of its intensified efforts, Meta raised its 2025 capital expenditure forecast to between $64 billion and $72 billion in April, a move clearly designed to position the company ahead of fierce competitors like OpenAI and Google.

With Prometheus and Hyperion on the horizon and an all-star AI team at the helm, Zuckerberg’s colossal bet on superintelligence signals not only Meta’s ambition—but also its willingness to bankroll the future of artificial intelligence at unprecedented scale. 

By fLEXI tEAM