Faculty Perceptions and Adaptation to AI Tools in Literature Assessment: A Study of Undergraduate English Programs

Abstract

The increasing use of Generative AI tools like ChatGPT in higher education has raised concerns about academic integrity and the efficacy of traditional assessment methods (AlAli & Wardat, 2024). This study, which was conducted in a leading non-state higher educational institute of Sri Lanka, explores how the faculty members in an undergraduate literature program perceive and adapt their assessment practices in response to the growing presence of AI tools. Semi-structured interviews with six lecturers teaching literature provided qualitative insights into their views on AI’s impact on student assessments, while document analysis of assignment guidelines, marking rubrics, and assessment policies helped to contextualize institutional responses to these challenges. The findings revealed a predominantly negative perception of lecturers towards ChatGPT, with faculty viewing it as a threat to the authenticity and originality of student submissions. However, revealed through document analysis it also highlights a gap in how faculty are adapting their assessment strategies to counter AI’s influence on student submissions. Many lecturers continue to rely on traditional essay-based assignments focused on evaluating students’ creative writing and analytical skills, without focusing on the process of writing or considering the potential use of AI tools. In response to concerns over AI-driven academic dishonesty, the faculty have shifted to in-class tests as a perceived solution. The study suggests that faculty reconsider their existing assessment practices and engage in developing more adaptive strategies. Faculty could collaborate to design an assessment toolkit for literature subjects, incorporating assignment templates, process-based assessment rubrics, and other resources to challenge AI-generated content while ensuring academic integrity.