Generative AI Literacy and Responsible Academic Writing Practices Among Bangladeshi University Students

Abstract

The rapid integration of generative artificial intelligence (GenAI) into higher education has transformed academic writing practices while creating new challenges in aspects of academic integrity, authorship, and ethical technology use. This study examines the level of generative AI literacy among university students in Bangladesh along with their level of responsible academic writing practices. Additionally, it explores the relationship between GenAI literacy and responsible writing behaviour and also aims to identify whether these practices differ across academic disciplines. The research was conducted by employing a quantitative cross-sectional survey, and 250 valid responses were collected from undergraduate students enrolled in different higher education institutions in Bangladesh through an anonymous online questionnaire. The survey instrument consisted of three sections: demographic information, generative AI literacy, and responsible academic writing practices. To analyze the data, descriptive and inferential statistical analyses were used with a view to examining students’ AI literacy levels, writing practices, relationships between variables, and disciplinary differences. The study findings indicate that generative AI has become a common component of students’ academic activities, with more than 71% of participants reporting frequent use of AI tools for purposes such as brainstorming, language improvement, summarization, and writing assistance. The research also revealed considerable awareness of AI capabilities, limitations, hallucinations, and ethical concerns. However, the results revealed that while most students engage in responsible AI-assisted writing practices, including fact verification, source checking, rewriting AI-generated content, and maintaining personal authorship, some students reported a need for clearer institutional guidelines and formal AI literacy training. Moreover, the findings also suggest a positive association between generative AI literacy and responsible academic writing practices, which is consistent with the relevance of Long and Magerko’s (2020) AI literacy framework for understanding ethical AI engagement. The study concludes by recommending that universities in Bangladesh should consider formal approaches and develop curriculum models for GenAI literacy to support critical evaluation, ethical AI use, and responsible academic writing practices.