An Authentic Assessment Framework For AI-Mediated Verbal Assessment in Commerce Education

Abstract

The rapid proliferation of generative artificial intelligence (GenAI) has challenged the validity of traditional written deliverables as authentic indicators of individual student capability in higher education. This paper presents a comprehensive, two-stage unified pedagogical framework designed to reclaim assessment integrity and elevate cognitive rigor within the economics discipline through an AI-Mediated Interactive Oral Assessment (AI-IOA) architecture. Shifting the evaluative focus entirely away from static written artifacts or external gamified simulations, the framework establishes a direct, dialogue-based verbal defence of threshold concepts across both microeconomics and macroeconomics. Drawing on Vygotskian scaffolding and authentic assessment philosophy, the model positions GenAI strictly as a constrained, Socratic interlocutor and multi-modal evidence-capture engine, while academic faculty retain final evaluative authority. We outline the theoretical model, provide a consistent blueprint mapping key economic principles to adaptive Socratic prompt workflows, and detail the technical implementation blueprints for integrating this framework natively within modern Learning Management Systems (LMS) environments.