UN University Report: AI Data Centers to Consume 945 TWh/Year by 2030 — Water and Land Costs Disproportionately Borne by Developing Nations

The United Nations University Institute for Water, Environment and Health (UNU-INWEH) released a major report on June 3, 2026, quantifying the environmental footprint of global AI infrastructure. Key findings: AI data centers will consume 945 terawatt-hours of electricity per year by 2030 — nearly triple the combined annual power consumption of Pakistan, Bangladesh, and Nigeria; cooling water demand from AI infrastructure will equal the basic annual domestic needs of 1.3 billion people; and the physical land footprint of AI data centers will exceed 14,500 square kilometers. The report warns that the transition to renewable energy for AI may reduce carbon emissions but sharply increase water consumption and land use, with the burden falling disproportionately on developing nations. Only 32 countries — representing 16% of all nations — currently host AI-specialized data centers, concentrating the economic benefits while environmental costs diffuse globally. The report calls for international AI infrastructure governance standards, mandatory environmental impact disclosure, and binding water-use efficiency targets for AI data centers, particularly in water-stressed regions. The findings are presented as a direct challenge to 'AI for Good' narratives that do not account for systemic environmental externalities.