Researchers and conservationists worldwide now have access to SA-FARI, an advanced artificial intelligence system designed to automatically detect, identify, and track nearly one hundred animal species in video footage. Developed by an international consortium led by ConservationX Labs and META, with pivotal contributions from the University of Bristol, the project represents a significant advancement in computational ecology and automated wildlife monitoring. Built upon META’s foundational Segment Anything Model 3, SA-FARI leverages vision-language processing to precisely segment and follow individual animals across video sequences. The system generates masklets, which are pixel-perfect outlines that track animals frame-by-frame through time. This technology isolates subjects from complex natural backgrounds, enabling accurate individual identification and behavioral analysis without manual intervention. The approach has the potential to eliminate thousands of hours of tedious video review traditionally required for camera trap surveys. To validate the system, researchers curated and annotated a comprehensive dataset comprising over eleven thousand wildlife videos captured in natural habitats. The entire dataset has been made publicly available to biologists, ecologists, and conservation practitioners to accelerate global research initiatives. The SA-FARI research paper is scheduled for presentation on June 6 at the Conference on Computer Vision and Pattern Recognition in Denver, United States. The work has been selected as an award candidate at the premier visual AI conference, marking the University of Bristol’s second consecutive year receiving this distinction. Tilo Burghardt, Professor of Computer Vision and Animal Biometrics at Bristol, emphasized that global conservation challenges demand collaborative technological solutions. Dr. Otto Brookes, Lecturer in AI and Animal Biometrics, noted that precise spatial and temporal animal localization is a fundamental prerequisite for monitoring behavioral responses to conservation interventions. The interdisciplinary project involved teams from the Hasso Plattner Institute, University of Oviedo, Osa Conservation, Senckenberg Museum of Natural History, Max Planck Institute for Evolutionary Anthropology, and Climate Corridors. First author Dante Wasmuht and senior author Didac Suris spearheaded the technical development, while Bristol researchers provided critical expertise in animal biometrics and conservation applications. Project leaders anticipate that SA-FARI will serve as an extensible platform for future ecological research. Planned enhancements include tracking animal body pose, estimating depth, and generating natural language descriptions of wildlife activities. By automating the most labor-intensive aspects of footage analysis, the system establishes a new standard for data-driven conservation monitoring, empowering researchers to scale their efforts and make faster, more informed decisions to protect vulnerable species.