Using AI could have a hidden cost: Water needed by 1.3 billion people

Artificial intelligence is becoming a bigger part of everyday life, helping people work, learn and create content faster than ever. But a new report from the United Nations University (UNU) warns that the technology's growing demand for water and electricity could create serious environmental challenges in the years ahead.

The study says the rapid expansion of AI could lead to a sharp rise in water use, energy consumption and land requirements by 2030, raising concerns about the hidden costs behind the technology.

AI's water demand could be massive

According to the latest report from the United Nations University Institute for Water, Environment and Health, data centres that power AI systems could use enough water by 2030 to meet the basic domestic needs of 1.3 billion people.

That figure is close to the number of people currently living across the entire African continent.

Researchers say water use has received far less attention than carbon emissions in discussions about AI. Large amounts of water are needed to cool servers and support the electricity generation required to run powerful AI models.

Electricity use set to surge

The report estimates that AI-driven data centres could consume 945 terawatt-hours of electricity every year by 2030.

That is nearly three times the combined annual electricity consumption of Pakistan, Bangladesh and Nigeria.

Researchers say the growing demand for computing power is putting increasing pressure on energy systems around the world.

AI relies on a large physical network

The UN-backed study argues that AI is not just a digital technology.

Behind every AI tool is a vast network of data centres, advanced computer chips, cooling systems and energy infrastructure. Building and operating this infrastructure requires significant amounts of land, water and electricity.

Everyday AI use has a bigger impact

One of the report's most striking findings is that using AI every day may consume more energy than training the models themselves.

Researchers estimate that daily interactions with AI account for around 80 to 90 per cent of total energy demand. Tasks such as generating images and videos use much more energy than simple text-based requests.

Call for responsible AI growth

The United Nations University and the UN are not calling for a halt to AI development.

Instead, they are urging governments, technology companies and users to adopt a more responsible approach that takes water, land and energy impacts into account alongside innovation.