Responsible AI: Environmental Monitoring and Sustainability

In the first of the 2020 Responsible AI series, director of the Scientific Responsibility, Human Rights, and Law Program of AAAS, Jessica Wyndham (AAAS), hosted AI experts Dr. Michael A. X. Izatt (Hitachi Vantara) and Professor Bistra Dilkina (University of Southern California) to discuss applications and limitations of AI.

They discussed how AI is being used for several environmental and sustainability applications such as satellite image processing for early detection of deforestation, wildfires or illegal activities in protected areas. However, the experts agreed that the significant environmental impact of AI’s heavy power consumption and the large amounts of data required for deep learning models are current limitations. They highlighted the need for AI models that can function with small amounts of data and that are more accessible in terms of computational resources. They also emphasized the importance of sharing AI models and data to advance environmental monitoring globally, especially in developing countries.