BEYOND THE WRITTEN TEST: RECONSIDERING ORAL ASSESSMENT IN FOREIGN LANGUAGE EDUCATION IN THE AGE OF GENERATIVE AI

Abstract

The rapid diffusion of Generative Artificial Intelligence (GenAI) is prompting a critical reconsideration of established assessment practices in foreign language education. This paper argues that a continued reliance on written assessment, increasingly vulnerable to the unauthorised use of AI-enabled technologies and personal devices, may undermine the validity of language assessment. Drawing on the principles underlying ALTE guidelines and the Common European Framework of Reference for Languages (CEFR), the article challenges the widespread assumption that standardised written testing necessarily provides a more objective measure of linguistic competence. It argues instead for a renewed centrality of oral performance, which enables teachers to observe learners’ linguistic resources and communicative competence in real time. Oral assessment, when supported by clearly defined criteria and appropriately designed tasks, offers a particularly direct means of assessing spontaneous language use, interaction and the negotiation of meaning. The paper also addresses practical and psychological concerns associated with oral assessment, particularly its implementation in large classes and its relationship with foreign language anxiety. It concludes that the emergence of GenAI should not simply be regarded as a threat to academic integrity, but as a catalyst for rethinking assessment practices and restoring greater importance to forms of evaluation that privilege authentic, synchronous and interpersonal language use.