Balancing innovation and integrity: Faculty perceptions of AI and generative AI in assessment

Abstract

The integration of Artificial Intelligence (AI) and Generative AI (GAI) into higher education assessment is reshaping debates, especially in business education, in teaching and learning. While AI supports data analysis, pattern recognition, and predictive modelling, GAI extends these capabilities to the automated generation of texts and assessment materials. Yet their role remains contested, raising questions of efficiency, ethics, and academic integrity. This study investigates faculty perceptions and practices regarding AI/GAI in assessment, with a focus on business disciplines. A national survey yielded 111 responses across institutions; 46.8% were social sciences faculty. The questionnaire examined current uses, perceived benefits and risks, and expectations for future adoption. Overall, 53.2% report using AI/GAI for assessment. Among adopters, applications include plagiarism detection (63.3%) and content generation (37.6%). Benefits include time efficiency (66.1%), greater objectivity (48.6%), and improved feedback quality (42.2%). However, concerns persist about reliability (62.4%), ethical implications (52.3%), loss of the human element (51.4%), and insufficient training (50.5%). Faculty remains divided on permissibility: 44.0% support conditional integration, 43.1% favor selective use, and 5.6% advocate complete prohibition. These patterns indicate cautious optimism. While faculty recognize AI/GAI’s transformative potential, training deficits, technical limitations, and unresolved ethical challenges impede widespread adoption. Business educators emphasize the tension between efficiency gains and safeguarding critical thinking, creativity, and fairness. This article offers one of the first empirical portraits of Portuguese assessment practices involving AI/GAI. It underscores the need for targeted training, ethical frameworks, and institutional policies that balance innovation and integrity, enabling responsible and effective assessment in higher education.

Ähnliche Arbeiten