Artificial Intelligence (AI) end-to-end: The Environmental Impact of the Full AI Lifecycle Needs to be Comprehensively Assessed - Issue Note

This issues note reviews the environmental implications associated with artificial intelligence across its full life cycle and examines both opportunities and risks for environmental sustainability. Prepared to inform governments, researchers, industry and the public, it synthesises recent literature to explain how artificial intelligence systems operate and how their software and hardware life cycles influence environmental outcomes. The document outlines stages including data preparation, model development, training, deployment and infrastructure production, highlighting associated resource demands. It identifies direct impacts such as energy, water and mineral consumption, emissions and electronic waste, alongside indirect and systemic effects linked to economic and behavioural changes. The note also discusses challenges in measuring these impacts and emphasises the need for improved metrics, reporting frameworks and research to support more sustainable development and deployment of artificial intelligence technologies.