The Uneven Distribution of AI's Environmental Impacts

The article discusses the environmental impacts of artificial intelligence (AI) technologies, particularly focusing on the significant energy and water consumption during the training process of AI models. It highlights that AI model training can lead to the evaporation of large amounts of freshwater into the atmosphere, exacerbating stress on freshwater resources. The uneven distribution of these environmental impacts across different regions and communities is emphasized, with a call for companies to manage the growing water and energy demands of their data centers responsibly. The authors suggest that flexibly deploying and managing AI computing across a network of geographically distributed data centers could help address the environmental inequality caused by AI technologies. The adoption of AI has the potential to address global challenges like climate change and drought mitigation, but concerns about the environmental impact of energy-intensive deep neural networks are growing. The article underscores the importance of prioritizing disadvantaged regions and equitably distributing the negative environmental impact of AI technologies.