Jakarta, ThedailyID — Artificial intelligence could place unprecedented pressure on global resources as demand for AI services continues to surge, according to a new report from the United Nations University Institute for Water, Environment and Health (UNU-INWEH).
Researchers warned that AI’s environmental footprint extends beyond carbon emissions. The technology increasingly consumes vast amounts of water, electricity, land, and raw materials.
The report estimates that by 2030, AI-related water consumption could exceed the amount of drinking water needed by the world’s population for more than 1.5 years. Data centers require large volumes of water for cooling systems. Power generation also contributes to rising water demand.
Researchers project that AI could consume nearly 3% of global electricity by the end of the decade. Annual demand from AI-powered data centers could reach 945 terawatt-hours. That figure exceeds the combined electricity use of several major developing countries.
Carbon emissions remain another growing concern. The report estimates AI-related emissions could reach nearly 400 million metric tons per year by 2030. That amount is roughly equal to the United Kingdom’s total emissions in 2025.
The study also highlights a rapidly expanding electronic waste problem. Researchers estimate AI infrastructure could generate around 2.5 million metric tons of e-waste annually by 2030.
Experts compared that volume to discarding about 250 Eiffel Towers worth of electronic equipment every year. Frequent upgrades to servers and AI processors contribute significantly to the problem.
Meanwhile, AI’s physical footprint continues to expand. Researchers estimate that AI-related infrastructure could occupy more than 14,500 square kilometers worldwide by 2030.
According to the report, public discussions often focus on emissions while overlooking other environmental costs. Researchers say policymakers and technology companies should also address water use, land consumption, mining activities, and electronic waste.
The report further highlights the Jevons Paradox. Greater efficiency does not always reduce resource use. Instead, lower costs can encourage wider adoption and increase overall consumption.
Researchers also noted that AI development remains concentrated in a handful of countries. The United States and China control most AI-specialized computing capacity. Many other nations have limited access to advanced AI infrastructure.
However, environmental impacts can spread far beyond those countries. Water stress, mining operations, and e-waste may affect regions that do not directly benefit from AI investments.
As governments and companies continue investing billions in artificial intelligence, researchers urge a more sustainable approach. They warn that unchecked growth could place increasing pressure on critical global resources during the coming decade.
