EVALUASI APLIKASI CHATGPT MENGGUNAKAN USABILITY TESTING.pdf

This study aims to evaluate the usability level of the ChatGPT application based on Nielsen’s five key usability dimensions: Learnability, Efficiency, Memorability, Errors, and Satisfaction. Using a quantitative approach through usability testing, data were collected via structured questionnaires from 55 ChatGPT users in Indonesia. The analysis was conducted using Partial Least Squares Structural Equation Modeling (PLS-SEM). The findings indicate that only the dimensions of <i>Satisfaction</i> and <i>Errors</i> have a statistically significant impact on overall usability. These results highlight that user satisfaction and minimizing errors are the most influential factors in enhancing the usability of AI-based conversational systems like ChatGPT.