Design Principles for Integrating Generative Artificial Intelligence in Knowledge Management

Recent advancements in Generative Artificial Intelligence (GenAI), particularly Large Language Models (LLMs), have garnered significant interest for their potential to enhance Knowledge Management (KM). Despite increasing adoption by organizations to improve the utilization of internal knowledge, a holistic framework for integrating this new technology remains absent. Through an Action Design Research (ADR) approach in collaboration with a public university, an LLM-powered chatbot was developed and evaluated for KM-related use cases, leading to the derivation of seven design principles: Data Privacy and Security by Design (1), emphasizing secure deployment environments and strict access controls; Scalable integration with Retrieval Augmented Generation (2), enabling efficient connection to organizational knowledge; Trust and Transparency by Design (3), incorporating source references and acknowledging system limitations; Data Quality by Design (4), ensuring up-to-date and well-structured knowledge bases; User-Centricity by Design (5), supporting effective prompting and feedback mechanisms; User Experience by Design (6), focusing on interface design and customizability; and Use-Case-Specificity by Design (7), tailoring systems to organizational context. This research provides actionable guidelines for integrating GenAI into KM processes.