UN warns AI boom could lead to massive water and energy consumption

The rapid expansion of artificial intelligence poses a significant environmental challenge, demanding vast quantities of water and electricity, according to a report by the United Nations University Institute for Water, Environment and Health (UNU-INWEH).

By 2030, AI is projected to consume 3% of the world's electricity production. This figure is nearly triple the combined annual electricity consumption of Pakistan, Bangladesh, and Nigeria. Furthermore, AI's energy demands are expected to generate emissions comparable to those of the United Kingdom. The report also highlights the immense water requirements for cooling AI infrastructure, estimating that it could consume more water annually than the entire world currently uses for drinking purposes.

The study estimates that annual water consumption linked to AI will reach 9.3 trillion liters by 2030. This amount could satisfy the yearly needs of over 1.3 billion people in Sub-Saharan Africa. The physical footprint of AI is also expanding, with data center surfaces expected to double the size of Jakarta, Indonesia's capital, by the end of the decade. The report also warns of the "Jevons paradox," where increased efficiency in resource use leads to a rise in overall consumption.

In 2025, data centers worldwide consumed 448 terawatt-hours of electricity, with AI accounting for approximately 20%. If data centers were considered a country, they would rank 11th globally in energy consumption, between Saudi Arabia and France. In 2025, they also generated 189 million tons of carbon dioxide emissions and used 4.5 trillion liters of water. The growth of AI is also expected to fuel a surge in e-waste, projected to reach 2.5 million tons annually by 2030. This waste could disproportionately affect lower-income countries with limited recycling capabilities, exacerbating social inequalities. The extraction of minerals required for building more data centers also raises environmental concerns and risks for local communities.

“

these findings are likely just the tip of the iceberg