UAE report: AI data centre water use to double by 2030

The latest:

Water consumption by data centres is projected to roughly double by 2030 to 9.32 trillion litres, according to a new UAE Foreign Ministry report on the intersection of artificial intelligence, water and energy. The report argues policymakers have largely overlooked what AI’s expansion means for strained resources, and calls for water, energy and digital infrastructure to be planned as one interconnected system.

Details:

• The report: Titled Crosscurrents: Navigating the AI-Water-Energy Nexus, the document was issued by the UAE Foreign Ministry. It maps water use across the entire AI value chain, covering semiconductor manufacturing, electricity generation and cooling infrastructure, rather than treating data centre consumption as the whole picture.
• The hidden cost: Indirect water use often exceeds direct consumption at the data centres themselves, according to the report. Power generation and chip production are identified as the two largest hidden draws, a finding that reframes how the water footprint of AI should be measured and regulated.
• The top official: Sheikh Abdullah bin Zayed, Deputy Prime Minister and Minister of Foreign Affairs, wrote in the executive summary that AI’s rise has shifted both the nature of the global water crisis and the ability to address it. He said governments and industry must manage the three systems together, and that used wisely the technology could help secure safe water access for all.
• The political limit: A key constraint on future AI growth is political, the report says, citing the United States, Malaysia, Chile and Ireland. Community concerns over water access and electricity demand are delaying or reshaping large data centre projects in those markets.
• The UAE case: The report points to domestic projects including MaiWater, which it says combines AI with renewable energy and water planning, covering advanced data centre cooling and strategic water storage. It is presented as an example of what integrated planning can deliver in a water-scarce state.
• Five recommendations: The report calls for international cooperation on responsible AI water use, integration of AI infrastructure into water and energy planning, mandatory transparent resource-use reporting, promotion of low-water and low-energy technologies, and embedding water and climate risk into corporate strategy.
• The core demand: “The priority is integrated governance: planning water, energy, infrastructure, and data as mutually dependent systems,” the report says. It does not specify which body would enforce such reporting requirements or set the standards.
• The UN warning: In January, the United Nations said parts of the world have entered water bankruptcy, a state of depletion that cannot realistically be reversed. The report positions AI’s resource demands against that baseline of already irreversible loss in some regions.

Between the lines:

The report’s framing shifts the AI water debate away from data centre cooling, where most scrutiny has landed, toward the power generation and semiconductor plants it identifies as the larger indirect draw. Its citation of four countries where community pushback is reshaping projects suggests the constraint on AI growth may arrive through local permitting fights before it arrives through physical water limits.

What’s next

The UAE and Senegal host the 2026 UN Water Conference in Abu Dhabi in December, which the report calls a premier opportunity to turn responsible AI development into a shared global agenda. Watch whether its transparency recommendations reach the agenda.