Pedagogical translanguaging with GenAI: From dilemmas to design

Abstract This article revisits the dilemmas of creativity–standardization, inclusion–fixity, and meaning–form, as proposed by Jeon et al. (2025), in relation to the intersection of generative artificial intelligence (GenAI) within translanguaging. While their contribution centers on a mechanism-level analysis of the input–output asymmetry of large language models, I adopt a pedagogical perspective and argue that these dilemmas remain pertinent in the context of pedagogical translanguaging. However, they may not always apply in the same way across classroom settings, particularly when standardization, fixity, and form are not inherently problematic and are instead intentionally calibrated by educators to foster creativity, inclusion, and meaning. I therefore recast the dilemmas as questions of classroom governance rather than properties of GenAI or translanguaging, and propose a reciprocal framing–reframing approach in which pedagogical translanguaging sets the terms of GenAI’s participation and GenAI, once introduced, presses those terms to be made explicit and auditable.