Generative Artificial Intelligence in Tertiary Level Education in Bangladesh: Practices, Benefits, Challenges, and Prospects

Aim/Purpose: This study aimed to investigate the potential of integrating Generative Artificial Intelligence (GenAI) in tertiary education. It examined current practices among teachers and learners regarding GenAI, as well as their perceptions of its benefits and challenges. Background: Higher education worldwide is seeing the increasing use of GenAI. However, its usage patterns and teachers’ and learners’ perceptions of its adoption are yet to be studied. The feasibility and viability of this emerging tool can be assessed by examining early usage patterns as predictors of formal adoption, as supported by the Technology Acceptance Model (TAM) and the Task-Technology Fit (TTF) frameworks. This study aims to fill that gap by examining both teachers’ and students’ practices and perceptions regarding various aspects of AI and its adoption in education. Methodology: A mixed-method approach was employed. Data were collected from 44 teachers and 186 students at Jashore University of Science and Technology through workshops and structured questionnaires based on the TAM and TTF frameworks. Quantitative data were analyzed with SPSS and MS Excel, while qualitative data were thematically analyzed. Contribution: This study contributes empirical evidence on the adoption of GenAI in a South Asian tertiary education context, enriching the body of knowledge on technology acceptance, digital pedagogy, and GenAI in education policy. By revealing the pictures of relevant variables of Generative Artificial Intelligence in Education (GenAIEd) in a unique context, such as Bangladesh, the findings have implications for similar situations. They can inform others about possible challenges and the usefulness of GenAIEd. Findings: Teachers and students are both moderately familiar with GenAI. The teachers primarily use it to prepare courses and materials, while students sporadically engage with GenAI, mainly for academic problem-solving, and they emphasize its role in personalized, learner-centered learning. GenAI familiarity is found to be a strong predictor of usage frequency. However, teachers express concerns about the reliability of GenAI, ethical implications, and the potential for deskilling. While the benefits and usefulness dominate, possible challenges and threats are marginally associated with the future adoption and use of GenAI. This finding is unique because, despite the overpowering ‘ease of use’ of the TAM model, ‘benefits or usefulness’ of the TTF model, challenges, and threats have been found as catalysts for GenAI adoption. Recommendations for Practitioners: Practitioners are to utilize GenAI to support, rather than replace, their teaching expertise. They should also encourage students to strike a balance between GenAI-assisted learning, critical thinking, and independent work. Furthermore, the institutions should introduce guidelines to ensure the ethical use of GenAI and academic integrity. Recommendation for Researchers: Researchers should explore the longitudinal effects of GenAI adoption on learning outcomes and skill development. They can also conduct comparative studies across different universities and disciplines. Investigating the role of GenAI in inclusive education and support for learners from disadvantaged backgrounds also demands research focus. Impact on Society: The findings highlight how GenAI can transform higher education in Bangladesh and similar contexts. It shows the importance of addressing the risks of overreliance and the unethical use of GenAI for effective learning. A balanced adoption could strengthen human–technology collaboration in education. On the other hand, it has revealed the aspects of GenAI, preferred by educators, that AI developers should consider. Future Research: Further studies should examine hybrid learning models that integrate GenAI with human expertise. Cross-cultural perspectives on GenAI in education remain another area of study. Furthermore, studies should be carried out to develop frameworks for maintaining academic authenticity while GenAI is being used in education.