Mapping business ecosystems using AI: A case study on energy transition

This study explores the application of artificial intelligence (AI) for mapping business ecosystems, particularly in the context of the energy transition, a phenomenon of global significance that requires the interaction of multiple industries, nations, technologies, innovations, and regulatory frameworks. An AI-driven method is introduced to process large-scale textual data and construct business ecosystems as social networks. By analyzing datasets from academic papers, online news, regulatory documents, and a combined dataset, the scalability and accuracy of these ecosystems were assessed. The validity of the method was tested through an expert survey involving 90 professionals across 9 countries, multiple industries, and diverse roles. Regression analysis revealed a statistically significant relationship between network centrality scores and expert assessments, demonstrating the robustness of the method. The combined dataset provided the highest model fit but required significant computational effort. In contrast, academic papers offered a more efficient and interpretable alternative, achieving similar accuracy with significantly lower processing demands and noise. The findings suggest that AI-driven methods can standardize, automate, and enhance real-world ecosystem mapping, reducing analyst subjectivity and improving scalability. This type of mapping is useful in both theory and practice because it enables scholars and decision-makers to identify structural dependencies, coordination patterns, and bottlenecks that may not be visible through manual analysis, thereby informing strategic positioning, diversification, scenario planning, and policy design. Future research should explore alternative AI models and validation methods, higher levels of functional aggregation, and longitudinal analyses to advance the study of energy ecosystems.