Mapping the boundaries of AI-supported innovation in vocational English teaching: A Q methodology study

Purpose As generative artificial intelligence (AI) is increasingly promoted as an innovation in language teaching, an important question is how teachers judge which AI-supported practices are pedagogically controllable, vocationally credible, and institutionally defensible. This study examines the viewpoint configurations through which English teachers in Chinese higher vocational colleges judge the boundaries of AI-supported innovation in vocational English teaching.Methodology The study employed Q methodology. Thirty-two teachers sorted 40 statements on an 11-point forced distribution, and by-person factor analysis was used to identify shared viewpoint configurations. Post-sort interviews with selected defining and confounded participants were conducted to support factor interpretation.Findings The retained three-factor solution explained 55.24% of the study variance and produced 30 defining sorts, with two confounded sorts and no non-significant sorts. The three configurations were interpreted as Workload-Sensitive Pragmatists, Authenticity-Oriented Gatekeepers, and Institutionally Anchored Adopters. Together, these configurations show three boundary-setting logics in teachers’ stated professional judgements: workload-sensitive support, authenticity-oriented filtering, and institutionally defensible adoption.Originality The study extends existing work on teachers’ selective technology use by specifying how AI-supported innovation is bounded through pedagogical controllability, vocational authenticity, and institutional accountability in vocational English teaching.Implications Professional development should move beyond generic AI training and help teachers make context-sensitive decisions about AI use, including low-risk material preparation, vocationally authentic tasks, and institutionally responsible classroom practices.