Designing AI-resilient assessment in higher education: a four-pillar conceptual framework

Abstract

Generative AI tools can produce polished academic text on demand, undermining the validity of assessments that treat written submissions as evidence of individual learning. Detection-based countermeasures have demonstrated variable accuracy and equity concerns. This paper does not report empirical outcomes or validation data. It proposes a framework for AI-resilient assessment that shifts evaluation from product quality to demonstrable reasoning, decision-making, and ownership of learning. The framework comprises four pillars: (1) process-based documentation; (2) oral defense integration; (3) authentic task design; and (4) transparent AI-use policies aligned with intended learning outcomes. These are operationalized through a rubric model, an oral-defense protocol, and an assessment vulnerability audit tool. Illustrated primarily through health sciences education, with wider relevance to professional and practice-oriented disciplines, the framework treats generative AI as a catalyst for re-examining how assessment evidence is generated rather than solely as an integrity threat. All proposed tools are design instruments for future empirical evaluation, not psychometrically validated instruments.