Deriving criteria for inclusive AI in museum environments: an abductive approach and case analysis

초록

Introduction – This study aims to derive and examine criteria for applying AI technologies in museum environments to secure inclusivity and to examine the structural factors that strengthen or constrain their effects. Methods – AI-driven inclusive services and products used in museums were classified into five domains: AI-based robots and kiosks, mobile services, online services, exhibits, and data analytics. A total of 20 cases were analyzed using qualitative and quantitative methods. Based on the Principles of Inclusive Design proposed by the Design Council, an abductive approach integrating deductive theory review and inductive case interpretation was employed to derive five criteria: Effortless Accessibility, Personal Adaptivity, User Centeredness, Universal Usability, and Operational Continuity. Results – The criterion-centered analysis indicates that AI-based inclusivity requires integrated design strategies addressing interaction structures, context-aware adaptivity, user agency, social integration, and operational strategy. The domain-based analysis further shows that strengths and limitations vary across AI technology domains, indicating the need for differentiated adoption strategies. Discussion and implications – The five criteria form an expanded framework reflecting museum environments where AI mediates visitor experiences. The standardized framework derived in this study enables systematic comparison across cases and can support the planning and evaluation of AI-based inclusive services and products in museums as well as other cultural institutions and public facilities, providing practical guidance for inclusive AI deployment under structural and technological constraints.