UN Secretary General asks tech companies to come clean on environmental cost of running AI data centres

Antonio Guterres has urged major AI companies to disclose the full environmental impact of their systems. The call sharpens scrutiny of the land, water and power demands behind the AI boom.

In Short

• AI systems rely on vast server networks housed in massive data centres
• These facilities need heavy electricity for servers, cooling and networking equipment
• Data centres also consume large volumes of water to prevent overheating

Artificial intelligence is becoming more powerful with each passing year, but the technology's rapid progress comes with a growing environmental cost. United Nations Secretary-General Antonio Guterres has called on major AI companies to publicly disclose the full environmental impact of their systems, arguing that the public deserves greater transparency about the resources required to power AI.

"AI data centres are hungry for land, water & power," Guterres said.

"I'm calling on every major AI company to publicly disclose the full environmental impact of its systems – as a matter of transparency."

"No more hidden costs. No more shifting the burden onto those least able to bear it. It is time to come clean."

Why is the UN concerned?

The growing concern stems from the massive data centres that power modern AI systems. While users interact with AI through simple chat windows and voice assistants, the technology relies on huge networks of servers housed in large data centres. These facilities are used to train AI models and generate responses whenever people use AI tools.

Running these centres requires enormous amounts of electricity to power servers, networking equipment, cooling systems, and other infrastructure. They also consume significant quantities of water to keep equipment from overheating. As AI usage continues to grow, so does the demand for these resources.

Public concerns are also growing

The environmental impact of data centres is becoming a concern for many communities. Some residents oppose the construction of data centres in their areas because of worries about electricity consumption, water usage, and land requirements.

A Reuters/Ipsos poll conducted in 2025 found that only about one-third of Americans approve of the rapid pace of data-centre construction supporting artificial intelligence. The survey also found that many people would oppose having a data centre built in their own community. The findings reflect broader public unease about the ongoing boom in AI infrastructure across the United States.

Tech companies are looking for solutions

Major technology companies are also aware of the growing concerns surrounding the energy and resource demands of AI. Google is reportedly pursuing an ambitious project known as Project Suncatcher, which aims to place AI data centres in space using solar-powered satellites equipped with the company's Tensor Processing Units (TPUs), specialised chips designed for artificial intelligence.

Elon Musk has also discussed the possibility of developing space-based data centres powered by solar satellites to meet the increasing computing demands of AI.

Meta is exploring a similar direction. The company recently announced a partnership aimed at collecting solar energy from space to help power its data centres and AI infrastructure. It has also partnered with another company to store renewable energy for extended periods.

- Ends