While AI promises solutions to climate change, its own environmental footprint is growing at an alarming rate. According to a 2025 study published in Nature Sustainability, AI systems could be responsible for 32.6 to 79.7 million tonnes of CO₂ emissions in 2025 alone — a carbon footprint comparable to that of a global city like New York (Nature Sustainability, 2025).
The energy consumption of data centers is surging. The International Energy Agency (IEA) estimated that data centers consumed 415 terawatt-hours (TWh) of electricity in 2024 — roughly 1.5% of global power demand. With AI adoption accelerating, the IEA forecasts this could more than double to 945 TWh by 2030. AI systems accounted for an estimated 15–20% of total data center electricity demand in 2024 (IEA, 2024).
The water footprint is equally concerning. Researchers at the University of California, Riverside found that training GPT-3 consumed approximately 700,000 liters of freshwater just for cooling. A single conversation of 20–50 questions with ChatGPT consumes roughly 500 ml of water (Li et al., 2023). The deployment of AI servers across the United States alone could generate an annual water footprint of 731 to 1,125 million cubic meters between 2024 and 2030 (VU Amsterdam, 2025).
Corporate emissions data confirms the trend. Google's total carbon emissions rose 48% over five years, while Microsoft's increased by 23.4% since 2020, largely driven by cloud computing and AI infrastructure. In 2022, Microsoft's AI rollout drove its largest year-over-year increase in both water and electricity consumption, raising total usage by one third (company sustainability reports, 2024).
A critical challenge is transparency. Major tech companies do not publish AI-specific figures on energy and water use, making independent assessment nearly impossible. As MIT researchers noted in January 2025: "We cannot manage what we cannot measure" (MIT News, 2025). Without mandatory disclosure requirements, the true environmental cost of the AI boom remains hidden.
Key Sources
- Li P., Yang J., Islam M.A., Ren S. (2023). Making AI Less Thirsty: Uncovering and Addressing the Secret Water Footprint of AI Models. arXiv:2304.03271
- Luccioni A.S., Viguier S., Ligozat A.L. (2023). Power Hungry Processing: Watts Driving the Cost of AI Deployment? ACM FAccT.
- International Energy Agency (2024). Electricity 2024: Analysis and forecast to 2026.
- Nature Sustainability (2025). Environmental impact and net-zero pathways for sustainable AI servers in the USA.