Meta is building a massive new data center named Hyperion to power its next-gen AI efforts, according to CEO Mark Zuckerberg in a recent Threads post. The facility will eventually generate five gigawatts of computing power to support Meta’s ambitious AI lab.
Zuckerberg mentioned that the size of the Hyperion would be big enough to cover most of Manhattan. A Meta spokesperson, Ashley Gabriel, told TechCrunch that Hyperion will be based in Louisiana, most likely in Richland Parish, where Meta has already announced a $10 billion investment. The plan is to have two gigawatts online by 2030, with the full five gigawatts coming online gradually in the years that follow.
READ: Congressman Ted Lieu’s mission to balance AI innovation and safety (May 9, 2025)
He further added that “Meta Superintelligence Labs will have industry-leading levels of compute and by far the greatest compute per researcher. I’m looking forward to working with the top researchers to advance the frontier!”
Meta CEO Zuckerberg also shared that company is gearing up to launch a massive 1-gigawatt AI supercluster, called Prometheus, in 2026. Once it goes live, it’ll put Meta among the first tech companies to operate an AI data center at that scale.
However, in response to this post on threads, several users have mentioned about the horrifying environmental changes that are going to happen because of this data center. One of the users stated, “ No animals, plants, or humans in *multiple* spaces the size of Manhattan?!What is it that AI has to do so badly? Remember DeepSeek?” While another user stated, “And whose drinking water will you be taking to cool these?”
Across the U.S., AI-powered data centers are putting a serious strain on local water resources and the impact is becoming harder to ignore. These centers, which need constant cooling to run massive AI models like ChatGPT and Gemini, often rely on evaporative cooling systems that consume millions of gallons of water each day.
READ: More people join Meta; Play AI’s ‘entire’ team to lead AI-driven voice solutions (July 14, 2025)
For instance, Google’s data center in The Dalles, Oregon draws so much water that it accounts for more than 25% of the city’s total usage, sparking concern among residents who rely on that same water for daily needs. Similarly, in West Des Moines, Iowa, Microsoft’s AI data campus used around 43 million liters per month during the summer, pushing the city to rethink how much water it can afford to allocate to tech companies without harming its own people.
What makes it worse is that many of these data centers are being built in drought-prone or water-stressed regions like parts of Arizona, Utah, and California where every drop matters. On the contrary, tech giants are still racing ahead with plans to build even more AI data centers across the U.S. Microsoft is developing multiple new campuses in states like Iowa and Wisconsin, while Google is expanding in Kansas, Nebraska, and Texas where water availability is already under pressure. Even Amazon has quietly filed plans for massive AI-focused infrastructure in Indiana and Oregon, where local water utilities are scrambling to figure out how to support the added demand.

