AI-aware assessment for the EFL classroom

Abstract

Generative AI has not made assessment impossible. It has exposed assessment that was never measuring much. Rather than banning AI tools, the EFL classroom needs assessment that protects validity, that keeps a learner's thinking visible, while building the AI literacy this generation will need. This article starts where the work starts, with classrooms co-constructing responsible-use guidelines, and then sets out three frameworks that fit together: DigComp 3.0, the European reference for digital competence, which now integrates AI and is built to inform assessment; the AILit Framework, its primary and secondary companion; and the AI Assessment Scale, a practical tool for deciding how much AI a task should allow. It closes with ten worked examples for lower and upper secondary English, two for each level of the scale, and a set of design principles.