Social media platforms have become a major facet of modern information systems, providing organizations real-time insights into stakeholder sentiments that can steer strategic decisions and corporate governance. This study explores how information systems-enabled sentiment analysis can be used to track reputational capital and foster stakeholder trust. Based on stakeholder theory and reputational capital models, we applied a hybrid AI method combining large language models with clustering techniques to analyze more than 36,000 social media posts gathered during the retirement of a well-known public figure.

Our analysis revealed distinct emotional groups such as positive, neutral, and negative reflecting emotions of admiration, uncertainty, and disappointment. These insights also bring out valuable actionable information for early detection of reputational risks, emphasising transparency in corporate reporting, and improving communication with stakeholders. Academically, this research advances the information systems discipline by presenting a method to transform unstructured social media data into  efficacious decision-support tools. For practitioners, it demonstrates how organizations can develop public accountability, stakeholder trust, and strengthen resilience in today’s transparent digital environment.