Harnessing Generative AI for Land Conflict Detection in the Brazilian Amazon
Abstract
This paper explores the use of high-resolution satellite data and multimodal large language models (LLMs) to analyze and monitor illegal mining operations in protected indigenous reserves in Brazil. We examine the role of these technologies in addressing the limitations of traditional remote sensing applications in political science and economics, particularly when studying phenomena that require high spatial and temporal accuracy. Using the Brazilian Amazon as a case study, we aim to establish an event study that assesses trends in illegal mining during and after pandemic-related policy shifts, highlighting the intersection of political changes and environmental impacts.
Type
Publication
Working paper
