What is the environmental footprint of artificial intelligence?

Artificial intelligence (AI) enabled technologies have the potential to support positive climate action whether through technology that models the Earth, to algorithms that make data centres more efficient. While this supports the green transition, the use of such AI systems raises concerns around sustainability particularly relating to the consumption of natural resources such as electricity and water as well as the carbon emissions produced. Alongside this, the rise of deep learning and large language models has dramatically increased the amount of compute capacity AI systems need. As the use of AI increases, governments and policy makers need to understand the environmental impacts of AI to make evidence-based decisions which leads to the question: what is AI's environmental footprint?

Through this panel, the OECD aims to explore this issue and launch the report "The AI footprint: measuring the environmental impacts of AI compute and applications". This report is informed by experts from the OECD AI Expert Group on AI Compute and Climate and the Global Partnership on AI, Responsible AI Working Group and examines existing measurement tools and the key challenges surrounding quantifying the positive and negative environmental impacts of training and deploying AI models and their applications. Participants will be able to discover this report and hear from experts in the sector.

Registration is required for this event and can be done on the OECD event page.