Speaker
Description
Large mixed-ability English classes often rely on one reading text and one set of tasks for all learners. This one-size-fits-all approach can leave developing readers overwhelmed and more proficient readers underchallenged. Generative AI offers new possibilities for rapidly adapting texts, scaffolding comprehension, and extending tasks; however, uncritical use may introduce distorted meanings, inappropriate language levels, weak questions, cultural bias, or inaccurate answer keys. This interactive workshop presents a human-in-the-loop approach in which AI supports differentiation, while teachers retain responsibility for pedagogical decisions and quality control.
Designed for secondary English teachers, the session follows a demonstration–practice–reflection sequence. Participants will begin with an Equity Scan to identify barriers in a common reading task. In the AI Differentiation Lab, they will transform one source text into three learning pathways: a scaffolded version for developing readers, a core version aligned with the target level, and an extended version for more proficient learners. They will then complete a Human-in-the-Loop Audit using a structured checklist to evaluate meaning preservation, CEFR appropriateness, cognitive demand, question validity, cultural relevance, and answer-key accuracy. Working in small groups, participants will revise one AI-generated version, conduct rapid peer review, and plan flexible classroom use without permanently labelling students by ability.
By the end of the workshop, participants will be able to generate, evaluate, and adapt AI-supported reading materials for diverse learners. They will leave with a reusable prompt template, a quality-control checklist, and a classroom-ready three-level reading task. The workshop argues that equitable AI integration is not about automating differentiation, but about combining technological speed with informed teacher judgement so that no reader is left behind.
