From Lab to Life · Essay

The Hidden Cost of AI: What Every Leader Needs to Know About Energy and Water

From Lab to Life: Translating complex research into practical insights

Published
12 August 2025
Author
Tali Režun
Topics
Leadership · Sustainability

A short leader-facing digest of the peer-reviewed COTRUGLI research on generative AI's resource footprint (B15). AI could double global data-centre electricity use to nearly 1,000 TWh by 2030. A 500-word ChatGPT email uses about 0.3 Wh of energy and 0.01 litres of water. The piece covers model-efficiency choices, regional grid risk, and practical mitigations such as task-specific models and off-peak training.