Higher education assessments in the age of GenAI: a systematic review

Abstract

Abstract Assessment in higher education has long centered on core principles such as validity, authenticity, and alignment with learning outcomes, alongside ongoing debates about how best to support academic integrity and meaningful feedback. The emergence of generative artificial intelligence (GenAI) is now placing renewed pressure on these established principles, prompting institutions to reconsider how assessment is designed, delivered, and evaluated. While a growing body of research has examined GenAI’s role in teaching and learning broadly, less synthesis exists on how it is reshaping assessment specifically, and how these shifts relate to long-standing assessment theory and practice. This systematic review provides an up-to-date examination of empirical research published between 2024 and 2026 on how GenAI is influencing assessment practices and supporting educators in assessing student learning. Using PRISMA, 72 studies were identified and analyzed. The findings show that research was conducted across 24 countries and 20 academic disciplines. Two overarching research purposes emerged: (1) investigating the use of GenAI in assessment and (2) examining perceptions of GenAI in assessment. Qualitative approaches were the most common research methodology. Three major assessment design approaches emerged in response to GenAI: (1) process orientation, (2) authenticity, and (3) oral and dialogical approaches. Five ways educators used GenAI across the assessment process were identified: (1) creating assessments, (2) practice for assessments, (3) grading, (4) transparency of learning, and (5) feedback. Together, these findings situate GenAI’s influence with the broader trajectory of assessment scholarship and provides researchers, educators, institutional leaders, and policymakers a clearer picture of how assessment practices are being negotiated in response to GenAI