NVIDIA has unveiled the Rubin generation of AI infrastructure, marking the world's first 100% liquid cooling system that eliminates water consumption entirely. According to NVIDIA, the new generation represents one of the biggest efficiency leaps in data center history, with every chip and networking component cooled entirely by liquid in a closed loop with no fans anywhere in the system. The cooling liquid can run at temperatures up to 45 degrees Celsius (113 degrees Fahrenheit), hotter than a typical hot tub, which allows facilities to rely on outdoor air rather than energy-intensive mechanical chillers for much of the year. As reported by NVIDIA, the approach effectively removes water from the cooling equation altogether, addressing one of the most critical concerns in AI data center operations. Ali Heydari, NVIDIA's director of data centre cooling and infrastructure, confirmed that "We have eliminated massive amounts of power usage and pretty much all water usage."
NVIDIA estimates that a 50-megawatt hyperscale facility could save more than $4 million annually in cooling-related energy and water costs by switching to liquid-cooled infrastructure. According to NVIDIA, the underlying technology is simpler than it sounds, using a liquid coolant made of water and propylene glycol that circulates in a closed loop without drawing in new water from external sources. Andrew Chien, a professor of computer science at the University of Chicago, explained to Fortune that the industry standard cooling temperature is currently 30 degrees Celsius, which demands considerably more air conditioning to maintain. "The thing that's exciting about what NVIDIA announced is it shows really what's possible in terms of pushing up this liquid input temperature to 45°C," Chien said, noting that this allows cooling without running HVAC units when outdoor temperatures are cool enough.
Elon Musk has backed NVIDIA's claim that data center water use is far smaller than critics suggest, providing crucial support for the company's environmental positioning. According to reports from The Information, Musk's endorsement comes as NVIDIA faces growing scrutiny over water consumption in AI facilities. The Tesla CEO, who runs large NVIDIA-powered clusters through xAI, has repeatedly praised NVIDIA's latest chips as the backbone of his AI projects, making his support particularly significant for the company's environmental messaging.
Eco Wave Power is developing wave-energy technology that uses NVIDIA AI infrastructure and digital twins to help convert ocean wave movement into clean electricity for AI data centers. According to NVIDIA, the company's system is designed to address power availability challenges as AI infrastructure demand increases rapidly. Seawater is roughly 800 times denser than air, allowing smaller wave-energy devices to generate significant power compared with wind systems. The technology uses NVIDIA Omniverse libraries to create digital twins that simulate wave activity, floater behavior, deployment layouts and operating conditions before installation, helping engineers test designs and reduce deployment risk. Once operational, NVIDIA accelerated computing and AI tools support predictive analytics, anomaly detection, environmental forecasting and maintenance planning to improve efficiency and reliability.
NVIDIA is not alone in the race to cut water use, with Microsoft announcing in August 2024 that its new data centres would stop using water for cooling altogether, a shift the company said would save more than 125 million litres of water per year per facility. The AI sector is confronting unprecedented energy challenges as perhaps half of data center projects due to start operating this year won't arrive on time, according to Currence, an AI analytics firm. Some 75 projects worth an estimated $130 billion were blocked or delayed in the first quarter alone, with trouble accessing electricity being a major factor alongside public resistance. More than 70% of Americans oppose data center construction in their area due to concerns about rising power bills, land and water usage, and AI's threat to jobs.
