Design principles for AI-powered metaverse language classrooms: A fuzzy Delphi study in higher education

Abstract

Artificial intelligence (AI) and metaverse technologies are reshaping second language education by enabling immersive, data driven, and collaborative learning. This study identified and prioritized design principles for AI powered metaverse language classrooms in higher education using the Fuzzy Delphi Method (FDM) with a panel of 16 domain experts. The panel evaluated a construct framework spanning pedagogy (heutagogy aligned), technology, assessment, ethics and implementation. Consensus thresholds (d ≤ 0.2; ≥75% agreement) and defuzzified scores were used to rank items for inclusion in a design guideline. Findings revealed strong expert consensus across all five domains. The highest priorities were Pedagogy (for immersive, authentic tasks) and Assessment & Analytics. Technology and Implementation & Support were also deemed essential but lower in priority. Among heutagogy capabilities, Share was ranked as the most critical element. The study yielded a validated, prioritized set of design principles to guide curriculum designers and instructors in deploying sustainable, learner centred metaverse language learning at scale.