Rethinking Achievement in English Medium Instruction: Assessment, epistemic access, and the challenge of Generative AI

This item comprises the presentation slides for the invited lecture “Rethinking Achievement in English Medium Instruction: Assessment, epistemic access, and the challenge of Generative AI”, delivered as part of the British Association of Applied Linguistics (BAAL) English Medium Instruction Special Interest Group (EMI SIG) Lecture Series in February 2026.The presentation critically examines how academic achievement has been conceptualised and operationalised in English Medium Instruction (EMI) research and practice. It highlights the longstanding tendency to conflate English language proficiency with academic achievement, and argues that such proficiency-led framings obscure the role of assessment design, curriculum alignment, and institutional structures in shaping learning outcomes.Drawing on applied linguistics, language assessment theory, and higher education research, the lecture foregrounds assessment as a central mediating mechanism between teaching, learning, and achievement in EMI contexts. It introduces the concept of epistemic access to examine how students’ opportunities to demonstrate disciplinary knowledge are shaped by linguistic demands embedded in assessment practices. The slides also address the implications of Generative Artificial Intelligence for EMI assessment, particularly in relation to validity, authorship, and equity, cautioning against the uncritical use of AI to reinforce existing assessment weaknesses.The presentation concludes by reframing achievement in EMI as a designed and institutionally produced outcome, rather than an individual learner attribute or a proxy for English proficiency. It outlines implications for EMI research, assessment redesign, and professional practice, with particular attention to fairness, responsibility, and multilingual knowledge-making.This material is archived as a citable scholarly and professional resource to support dissemination, teaching, and further research on EMI, assessment, and responsible engagement with Generative AI in higher education.