Is generative artificial intelligence (GenAI) trustable among LIS students? A study of ChatGPT using the stimuli-organism-response framework

Abstract

Purpose This study aims to identify the relationship of stimuli [learning value (LV), information accuracy (IA), perceived credibility (PC) and perceived intelligence (PI) of ChatGPT] with organism (trust and attitude towards ChatGPT) and response (intention to use ChatGPT) using the stimulus-organism-response framework. Design/methodology/approach The research was carried out using the quantitative method. The population of the study consisted of library and information science (LIS) students of HEC-recognised public sector universities situated in Punjab, Pakistan. Two LIS schools from two universities were randomly selected, and data was collected from ADP/BS, MPhil and PhD students through the convenient sampling technique. The questionnaire was developed through previous studies, and data was collected online using Google Forms. Data was analysed using SmartPLS software. Findings Findings revealed a significant effect of LV, IA, PC and PI on trust. There was also a significant effect of LV, PC and PI on attitude towards ChatGPT. However, IA did not significantly affect attitude. The influence of trust and attitude towards ChatGPT on intention to use ChatGPT was also significant and positive. Originality/value This study has significant implications theoretically and practically. Theoretically, current research provides a pathway for researchers to extend the proposed model for ChatGPT and other generative tools for measuring the trust of users. Practically, this research provides insights for information technology experts and AI system developers to consider LV, IA, PC and PI to enhance user trust towards AI generative tools.

Ähnliche Arbeiten