Strategies for e-Assessments in the Era of Generative Artificial Intelligence

Abstract

The rapid advancement of generative artificial intelligence (AI), particularly tools like ChatGPT, is reshaping educational landscapes by enabling students to generate responses that closely mimic human-written answers. This development presents both opportunities and challenges for e-assessments, especially concerning academic integrity and the authenticity of student learning outcomes. Traditional assessment methods, which often emphasize memorization and standardized testing, are proving insufficient in this new context, as they may not effectively measure higher-order skills like critical thinking, creativity, and problem-solving. This study employs a systematic literature review (SLR) to investigate adaptive e-assessment strategies in higher education that address the integrity challenges posed by generative AI while supporting meaningful learning. Through an in-depth analysis of recent literature on e-assessment practices and AI integration, this study identifies key adaptive strategies such as randomized questioning, project-based assessments, open-book exams, and AI-enhanced plagiarism detection. The findings reveal that while generative AI complicates the assessment process, it also provides an impetus for rethinking assessment design in ways that promote application-based knowledge and discourage cheating. By advocating for a shift towards assessments that evaluate critical skills rather than rote knowledge, this study proposes a framework that can support educators in creating robust, integrity-focused e-assessments. This research contributes to the evolving discourse on educational assessment, offering practical recommendations for institutions aiming to balance the benefits of AI-enhanced learning with the need for fair and accurate assessments.