Purpose: Recent advancements in artificial intelligence, particularly Large Language Models, offer new capacities for enhancing and refining Knowledge Management processes within organizations. This study elucidates language model applications within the fourth-generation Knowledge Management framework, covering the phases of creation and acquisition, storage and organization, sharing and dissemination, application and utilization, and evaluation and monitoring.



Methodology: This study employs a systematic literature review combined with thematic analysis. Following the PRISMA framework, data were collected from Web of Science, Scopus, and Google Scholar databases for the period 2023–2025. After a rigorous screening process, 18 articles were selected for final analysis from an initial pool of 108 documents. Data analysis was conducted using ATLAS.ti (version 24). To assess inter-coder reliability and account for chance agreement, 20% of the sample articles were recorded by a second coder, yielding a Cohen’s Kappa coefficient of 0.82. Furthermore, the content validity of the extracted codes was verified by three experts, with CVR and CVI indices calculated and confirmed.



Results: Findings demonstrate that language models play a significant role in KM processes, identifying 26 core applications categorized into five primary dimensions: Knowledge Creation and Acquisition (6 applications), Storage and Organization (5), Sharing and Dissemination (5), Application and Utilization (5), and Evaluation and Monitoring (5).



Conclusion: Achieving full potential of fourth-generation Knowledge Management requires more than reliance on language models alone. Instead, the systematic and integrated synergy between these models and structured knowledge systems, specifically Knowledge Graphs and Retrieval-Augmented Generation (RAG) architecture, is essential for driving profound organizational transformation.