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

2026-06-04 · Source: Artificial intelligence (AI) – The Conversation · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Environmental Science & Earth Systems, Public Policy & Governance · Depth: Fundamental Awareness, quick

Summary

A new United Nations report quantifies the environmental costs of artificial intelligence, challenging the notion that future efficiency gains will reduce resource demand. The report estimates that by 2030, AI's energy use could double to consume 3% of the world's electricity, produce emissions equivalent to the UK, and deplete more water for cooling than the global population's annual drinking water needs. This projection is based on the "Jevons paradox," where efficiency improvements lead to increased overall consumption due to expanded use and lower costs. The report also highlights structural inequity, with only 32 nations hosting AI cloud infrastructure, 90% of which is in the US and China, potentially widening the digital divide and disproportionately burdening developing nations with mineral extraction and e-waste. It proposes a roadmap for responsible AI, emphasizing transparency, efficiency by design, equity, lifecycle responsibility, global cooperation, and sustainable use.

Key takeaway

For policymakers and AI strategists developing national AI plans, your "light touch" regulatory approach risks overlooking the significant environmental costs of AI. You should integrate mandatory environmental disclosures into AI development frameworks, considering both model and task-level impacts. Proactively incorporate projected AI demand into national climate and energy planning to avoid exacerbating resource depletion and widening the digital divide.

Key insights

AI's increasing efficiency risks higher overall resource consumption due to the Jevons paradox, demanding urgent environmental governance.

Principles

• Efficiency gains can increase total resource consumption.
• AI governance needs full value-chain responsibility.
• Environmental disclosures are crucial for AI development.

Method

The report lays out a roadmap for responsible AI based on transparency, efficiency by design, equity, lifecycle responsibility, global cooperation, and sustainable use.

In practice

• Incorporate projected AI demand in climate planning.
• Make environmental disclosures routine for AI models.
• Rethink AI innovation for sustainable tech future.

Topics

• AI Environmental Impact
• Jevons Paradox
• Data Center Energy
• Responsible AI Governance
• Digital Divide
• Resource Consumption

Best for: Investor, CTO, VP of Engineering/Data, AI Ethicist, Policy Maker, Tech Journalist

Related on AIssential

• UN report warns AI could soon use 3% of world's electricity and more water than we need to drink — AI's environmental footprint is projected to grow significantly due to the Jevons paradox, demanding urgent responsible governance.
• UNU-INWEH: By 2030, global data-centre electricity use could nearly double to 945 TWh, with water impacts comparable to the basic annual domestic needs of 1.3 billion people in Sub-Saharan Africa. — AI's environmental footprint extends beyond carbon, encompassing vast electricity, water, land, and e-waste, driven largely by inference.
• AI Was Born Wrong — AI's current trajectory, driven by profit, undermines education, strains resources, and necessitates government intervention.
• AI laws overlook environmental damage – here’s what needs to change — AI's substantial environmental footprint is largely unaddressed by current global regulatory efforts.
• Anthropic promises to cover consumer electricity costs from new data center construction — AI companies are pledging to cover data center electricity costs amid growing regulatory and public pressure.
• Big Tech Will Not Save Us From the Climate Crisis — Big Tech's AI-driven data center expansion threatens climate goals and local environments through immense energy and water consumption.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Artificial intelligence (AI) – The Conversation.