The general perception is that the digital economy is environmentally friendly. This misconception, however, stems more from a combination of false perceptions and a lack of information than from reality. Inria, a French research institute specializing in digital sciences, cites forecasts indicating that by 2030, half of the world’s electricity consumption will be attributable to computing facilities.

Within this trend, the development of artificial intelligence—a technology that is, moreover, essential for the deployment of renewable energy—is a major factor. Inria cites studies indicating that the design and training of state-of-the-art machine learning models produce the same amount of CO2 as six medium-sized vehicles over their lifetime.

In this context, and with the participation of Inria, six institutions from five countries—including Chile—have begun working on two projects: one aimed at developing environmentally sustainable machine learning technology (GreenAI), coordinated by Inria Chile, and another aimed at developing Sustainable Artificial Intelligence (SusAIN). Researchers from Inria in France and Chile, the Center for Mathematical Modeling at the University of Chile, the University of Asunción in Paraguay, the National Laboratory for Scientific Computing in Brazil, and the University of the Republic in Uruguay are participating in the projects.