Sustainable AI is a Myth

This open access book examines the real-world consequences of Artificial Intelligence (AI) and Machine Learning (ML) development and use on a global scale, including the detrimental impact of AI and ML on the environment. More broadly, it shows how we are living within the myth of the "omnicompetence of AI/ML," in which AI can supposedly solve the world's most complicated problems without considering how deeply unsustainable the development, production and use of AI/ML actually is.

In recent years, the term 'Sustainable AI' has come into vogue to suggest using AI/ML to achieve and/or accelerate the United Nations Sustainable Development Goals (SDGs), including climate action (SDG 15), ending poverty (SDG 1), and gender equality (SDG 5) among others. By exploring the ways in which Sustainable AI is a myth, this book invites readers to imagine AI/ML as broken in the sense that there are serious societal and environmental concerns related to the design and development of AI/ML that demand repair. Scholars, students, professionals, policymakers, and the public learn about the concept of 'Sustainable AI' alongside the breadth of environmental and social costs associated with AI/ML development. Readers are left with a desire to, and the tools to, imagine a sustainable future with AI/ML.