UN report warns AI could soon use 3% of world’s electricity and more water than we need to drink

# UN Report: AI's Energy and Water Demands Growing Rapidly

A United Nations report warns that artificial intelligence systems could consume 3 percent of global electricity within years, with water consumption rising to levels exceeding human drinking water needs worldwide.

The energy demand stems from training and running large language models and other AI systems. Data centers powering these models require significant cooling, which drives water consumption. As AI becomes cheaper and more accessible, organizations will deploy it more widely, offsetting any efficiency improvements the technology itself delivers.

This pattern mirrors historical technology adoption. When computing became cheaper, it did not reduce overall energy use. Instead, people found more applications for it. AI follows the same trajectory. Lower costs mean more companies train larger models, run more inference tasks, and create new AI-dependent services. Each deployment adds to the total energy footprint.

The UN findings matter to educators, students, and parents because energy and water costs shape technology access and affordability. Schools and universities already compete for limited IT budgets. Rising infrastructure costs for AI systems could force difficult choices between deploying educational AI tools and funding other programs.

The report also highlights environmental equity concerns. Data centers concentrate in regions with cheap electricity and water access, often in areas already facing resource scarcity. Training a single large language model can consume millions of gallons of water and gigawatt-hours of electricity.

Some companies have pledged to improve efficiency, but pledges alone have not reversed consumption trends in other sectors. The tech industry continues expanding AI capabilities without clear limits on resource use.

Policymakers face pressure to establish standards for AI energy and water consumption, similar to regulations around carbon emissions. Educational institutions increasingly rely on AI for grading, tutoring, and administrative tasks, making energy accountability part of campus sustainability strategies.

The UN report underscores a fundamental tension: AI adoption accelerates while the environmental costs remain externalized and largely invisible to end users