AI-driven classroom transformation and the algorithmic challenges to educational equity and teacher autonomy

This study examines how university teachers perceive, interpret and negotiate AI-driven educational platforms in relation to algorithmic fairness, educational equity and professional autonomy. A sequential explanatory mixed-methods design was employed, involving the collection of survey data from 300 teachers across six universities in Chengdu, China, followed by semi-structured interviews with 30 purposively selected participants. The quantitative findings show that urban teachers and frequent AI users reported higher perceived fairness. However, many teachers lacked a practical understanding of how AI systems generate recommendations or feedback. Qualitative thematic coding and NLP-supported semantic mapping revealed additional concerns relating to algorithmic opacity, role ambiguity, training deficits and the “toolification” of teachers’ pedagogical judgement. The findings distinguish between “equity-by-design” and “equity-by-implementation”, demonstrating that the fair integration of AI requires transparent platform design and context-sensitive institutional support. The study emphasises the pivotal role of teachers as mediators, co-designers, and governance actors in AI-enhanced learning environments.