Elon Musk endorsed NVIDIA's environmental manifesto: the myth of AI's water scarcity debunked
Large technology corporations are increasingly coming under fire for their colossal consumption of natural resources, primarily electricity and fresh water. However, NVIDIA's management firmly refutes these accusations, and this time its position has been publicly supported by Elon Musk. The debate about the environmental footprint of cutting-edge technologies is reaching a fundamentally new level.
Real Numbers: How Much Water Do Data Centers Use?
According to data from the Manhattan Institute for March 2026, cited by NVIDIA, data centers consume only about 0.2% of all fresh water in the United States. Moreover, the lion's share of this volume comes not from direct cooling, but from indirect consumption — the generation of electricity that powers the servers.
NVIDIA's key argument is the revolution in cooling systems. Modern liquid cooling operates at a stable temperature of 45 degrees Celsius. A sealed, closed-loop system allows AI complexes in northern latitudes to fully switch to economical dry radiators. This eliminates the need for bulky evaporative towers, which were the standard for infrastructure of previous generations.
The company claims that its Blackwell platform is 300 times more water-efficient than traditional air cooling. Since maintaining optimal temperature consumes up to 40% of a data center's total energy, this innovative approach fundamentally changes the rules of global competition in the AI market.
Musk's Support and Hidden Risks
Elon Musk, whose company xAI uses large clusters based on NVIDIA solutions, has repeatedly called the latest chips a key element of his AI projects. His public endorsement significantly strengthens the partners' position. Developers aim to completely dispel the popular myth that neural networks deplete regional water supplies. The cooling system is an absolutely sealed closed loop: the liquid constantly circulates inside the equipment, so servers do not require regular replenishment from the water supply.
"NVIDIA's DSX reference architecture for AI centers does not consume water. We have managed to minimize energy consumption and almost completely eliminate the use of water," stated Ali Heydari, head of data center cooling and infrastructure at NVIDIA.
Limitations and Non-Obvious Risks
However, the loud claims of zero water consumption only concern the direct cooling loop. Indirect water consumption (for electricity generation) is rapidly increasing amid the large-scale deployment of neural networks. Analysts at the Berkeley Lab predict that direct water consumption by data centers will jump to 38–73 billion gallons by 2028, compared to 17.4 billion gallons in 2023.
The efficiency of dry radiators is directly tied to climate: it is excellent in cool states but drops sharply in arid desert zones. A striking example is the giant xAI computing complex, Memphis Colossus, which daily drew up to 1.3 million gallons of clean drinking water from underground sources. Additionally, the company launched dozens of gas turbines before obtaining official permits, sparking outrage among local residents and leading to high-profile lawsuits.
Cryptalist Expert Opinion: NVIDIA's environmental manifesto is a strong PR move, backed by real technological achievements. However, the AI industry is on the verge of a "water paradox": local innovations in cooling do not solve the problem of indirect consumption and regional imbalance. Future development will entirely depend on the stringency of environmental regulation by authorities, and here, Musk's support could play both a positive and negative role, creating an illusion of a complete solution where one does not yet exist.