An LLM-based pipeline to semi-automate the assessment of Invasive Alien Species impacts
The introduction and spread of invasive alien species (IAS) have become a significant concern worldwide, disrupting environmental balance, economic stability, and human health across various activity domains. Effective management of IAS requires a cross-sectoral approach and identification of direct and indirect effects on multiple activity domains. This study explores the potential of artificial intelligence (AI) to support automatic literature reviews and to evaluate IAS impacts. The study analysed 498 IAS impacting 27 activity domains and demonstrated the potential for semi-automation to enhance efficiency, the importance of expert involvement, and the value of AI in extracting contextualized information. The study highlights the feasibility of replicating the assessment on IAS using AI-powered approaches but also emphasizes the need for careful consideration of data quality, algorithmic bias, and the role of subject matter experts in validation and refinement. While we were able to successfully replicate the analysis, the process also revealed technical limitations that should be considered when applying this approach in practice.
Vazquez Torres E, Magliozzi C, Caivano A, Cardoso AC.
Publications Office of the European Union, Luxembourg, 2026, doi:10.2760/2865132