The evolution of Large Language Models (e.g. GPT-4) in the modern data-driven business contexts has opened up new perspectives in optimizing operations and managing information. This study introduces the Automated Semantic Taxonomy Enrichment Methodology (ASTEM), a novel framework utilizing GPT-4 to enhance the semantic richness of corporate taxonomies. ASTEM integrates advanced prompt engineering and iterative evaluation to generate contextually relevant taxonomy definitions. A case study carried out in a large company operating in the aerospace sector provides a practical perspective on the methodology effectiveness, demonstrating its crucial role in filling information gaps and establishing relevant semantic connections. This study demonstrates the potential of leveraging artificial intelligence to automate complex intellectual processes and suggests directions for future research in expanding its application across different industrial domains.