A new artificial intelligence model developed by researchers at the University of Maine and Oregon State University can assess extinction risks for over 10,000 freshwater fish species worldwide by analyzing 52 factors. The tool aims to help conservationists act before species become endangered, offering a proactive approach to protecting biodiversity. The model, led by University of Maine assistant professor Christina Murphy, identifies threats such as dam construction, water withdrawal, habitat destruction, pollution, invasive species, and socioeconomic pressures. By leveraging publicly available data from 12 sources—including the International Union for Conservation of Nature—the AI system can detect patterns that predict which species are most at risk. The research, published in Nature Communications, shows that while nearly one-third of freshwater fish species are currently threatened with extinction, many of them could still be saved if conservation efforts are implemented early. Species like Maine’s Arctic Char (Salvelinus alpinus) and other char populations around the globe are among those that could benefit from timely intervention. Unlike traditional assessments that focus on what’s going wrong, the model also identifies what’s working. It reveals consistent ecological, environmental, and socioeconomic conditions that support fish survival, allowing conservationists to replicate successful strategies across multiple species and regions. Co-author J. Andres Olivos, a postdoctoral researcher at Oregon State University, explained that the findings reflect a key insight: just as human health has predictable signs of well-being, freshwater fish species in safe conditions tend to share common, stable characteristics. In contrast, extinction risks arise from a wide variety of combinations of threats. The AI was trained on millions of complex, nonlinear relationships between species and their environments. It was validated against existing conservation assessments, ensuring reliability and accuracy. Users can input data on specific species to see which risks are present and whether those threats could escalate in the future. Ivan Arsmendi, an associate professor at Oregon State University, emphasized the model’s potential for early action. “People often wait until a species is already in crisis before stepping in. This tool lets decision-makers deploy resources proactively, before it’s too late,” he said. Murphy began the project in 2020 as a postdoctoral researcher at Oregon State, working alongside Arsmendi and Olivos. The team collaborated with scientists from the U.S. Geological Survey, the U.S. Forest Service, and the University of Girona in Spain. The model is designed not only for fish but could be adapted to protect birds, trees, and other wildlife, offering a scalable solution for global conservation planning.