Generative AI in Academic Writing and Discourse: A User Perspective
Sujoy Sen, Anshul Agrawal, Sanjeev Kadam, Shiladitya Verma · 2025
This study examines user perceptions of Generative AI (GenAI) technologies like ChatGPT in academic writing, focusing on engagement, motives, benefits, challenges, and integration. Gen AI tools like ChatGPT is a transformer trained in two stages: unsupervised pre-training and supervised fine tuning, to predict the next most probable word in a sequence to generate expected content. A survey of 206 respondents from diverse backgrounds identified four key roles of GenAI: facilitator of thoughts, reliable content creator, supportive brainstorming partner, and content augmenter. SPSS Version 20 was used to analyze data validity, reliability, normality, clustering, and hypothesis testing. Survey data was processed using principal component analysis with varimax rotation. The hypothesis testing was carried out using the non-parametric Kruskal-Wallis test with the help of p-significance. Statistical tests revealed no significant linkage between user qualification or category and GenAI usage for academic purposes. However, trust in GenAI's accuracy showed a gender bias, with males more likely to trust its outcomes than females. Interestingly, despite age often influencing technology adoption, no generational differences were found in GenAI usage. It suggests that its ease of use and operability make it accessible across age groups. The findings highlight the broad acceptance of GenAI in academia while underscoring trust-related gender differences.