This study aims to determine the effectiveness of an artificial intelligence (AI)-based chatbot compared to conventional methods in library information services at SMAN 9 Bandar Lampung. The research method uses an experimental quantitative approach with an Intact-Group Comparison design. The study population consisted of 1,102 students, with a sample of 92 respondents selected using the Slovin formula. The experimental group used a WhatsApp Business-based AI chatbot, while the control group used conventional services. Data were collected through a 1–5 Likert scale questionnaire that was valid and reliable (Cronbach's Alpha = 0.786). Prerequisite tests using Shapiro-Wilk and Levene's Test showed that the data were normally distributed and homogeneous. The results of the independent sample T-test showed t = 7.285 with a significance of 0.000 (p < 0.05), so H₁ was accepted. Descriptive analysis showed that the experimental group (M = 48.30) achieved the "effective" category for all indicators, while the control group (M = 38.65) was in the "neutral" category for the indicators of information speed and service availability without time constraints. In conclusion, the use of AI-based chatbots through WhatsApp Business has proven to be more effective than conventional services in improving the quality of library information services.
Copyrights © 2026