Key points are not available for this paper at this time.
Biological invasions are a primary driver of global biodiversity loss and impose socio-economic costs. While the application of Artificial Intelligence and Machine Learning in invasion science has grown in recent years, a significant disconnect remains between advanced computational capabilities and on-the-ground management and policy implementation in biological invasions. This mini-review synthesises current trends in AI applications in invasion science and ecology. The results reveal a landscape dominated by fundamental research and species distribution modelling, with fewer studies focusing on tools for rapid response and control. Critical barriers to integration are identified, including data heterogeneity, the “black box” nature of deep learning, and the lack of standardised training data for management outcomes. Finally, a “Nexus” framework is proposed to align AI development with the invasion policy cycle, emphasising explainable AI, multimodal data integration, and the inclusion of Traditional Ecological Knowledge to transform predictive algorithms into robotic treatment or physical action and actionable Decision Support Systems.
Al-Khawand et al. (Wed,) studied this question.