Artificial intelligence (AI) holds a rapidly expanding – but as yet not fully known – potential for advancing climate mitigation, adaptation, and ecosystem restoration goals. But bias from data and algorithms can deepen existing inequities, and the speed of AI’s evolution is raising ethical questions about who it will benefit, and who may be left behind.
CIFOR-ICRAF is looking at how we can integrate AI into our science in reliable, safe and equitable ways.
For example, by integrating diverse datasets and analytical tools powered by machine learning and AI, the Global Tree Knowledge Platform helps users – from policymakers to smallholders – navigate complex landscapes to make informed decisions about planting the Right Tree in the Right Place. The Land Degradation Surveillance Framework (LDSF) is a comprehensive method for assessing soil and land health, and is benefitting from the exploration of new and advanced data analytics. And the Regreening Africa app combines AI and citizen science to serve hundreds of thousands of farmers and reverse land degradation across eight countries in sub-Saharan Africa.
Other potential uses include AI-driven analysis and virtual reality to better predict which tree crops can grow in various conditions, efficient identification of wildlife species in camera traps and the prediction of animal and poacher behaviours, and the collation, analysis, and distribution of data to promote and amplify grassroots initiatives, Indigenous knowledge, local wisdom, and community-led actions. Despite its potential, AI uses massive amounts of energy and data; this poses ethical issues and hampers IT companies’ ability to reach their ‘green’ targets. In view of this and our commitment to sustainability, we are working to build on the benefits of AI while supporting efforts to minimize the environmental impacts.
