Knowledge Management Systems in the Age of Generative AI: Rethinking Knowledge Processes
The rapid advancement of artificial intelligence is fundamentally reshaping the architecture and logic of Knowledge Management Systems (KMS). Traditionally, KMS have been designed around repositories, taxonomies, and retrieval mechanisms for storing and redistributing explicit knowledge. However, in the AI era – particularly with the emergence of generative models – the role of KMS extends beyond storage and retrieval, towards active participation in knowledge processing and knowledge creation. This paper examines the evolution of KMS in the AI era through the lens of the SECI model (Socialisation, Externalisation, Combination, Internalization). It argues that AI tools can introduce a new operational dynamic within each SECI phase if properly addressed. In the externalisation process, AI systems can facilitate the conversion of tacit and loosely articulated insights into structured representations. In the combination phase, machine learning models may enable pattern discovery and synthesis across heterogeneous knowledge sources. During internalisation, AI-powered assistants can support experiential learning by contextualising and personalising information. Most importantly, socialisation can be augmented through collaborative AI-mediated environments that enhance collective intelligence, reshaping how shared meaning is constructed. The study critically explores how AI-enhanced KMS can transform from passive infrastructures to evolve into adaptive cognitive systems supporting KM. While AI may increase speed, scalability, and pattern recognition, it also introduces epistemological risks – such as bias propagation, over-automation, and erosion of human judgment. The paper discusses how organisations can mitigate these risks while developing resilient and adaptive KM practices. Adopting a conceptual and integrative approach, the research analyses current technological capabilities and conceptual KM frameworks. This research proposes an updated perspective on KMS as a hybrid socio-technical ecosystem. In such systems, AI tools do not replace human knowledge actors but extend their cognitive and organisational capacities. The paper concludes that the future of KMS lies not in more sophisticated repositories, but in intelligent systems capable of dynamic codification, contextual reasoning, and continuous organisational learning, redefining the balance between human and machine agency in organisational knowledge processes.