Traditional and Artificial Intelligence-based Assessment in Foreign Universities

Abstract

The rapid advancement of artificial intelligence (AI) is fundamentally reshaping how student learning is assessed in higher education institutions worldwide. This article provides a comparative analysis of traditional assessment methods (written examinations, essays, oral defenses, project-based evaluation) and AI-based assessment tools (automated grading systems, large language model feedback, AI-proctoring) as applied in foreign universities. Drawing on recent empirical studies, systematic reviews, and institutional practices from universities such as Cornell, Purdue, UC San Diego, and Gothenburg, the study identifies the strengths, limitations, and academic-integrity implications of both approaches. The findings suggest that while AI-based assessment offers clear advantages in speed, scalability, and immediacy of feedback, human evaluation retains an edge in assessing complex, creative, and context-dependent work. The article concludes that a hybrid model — combining AI-assisted tools with human oversight — represents the most promising direction for future assessment practice.