Reference Service and Inquiries-Answering Using ChatGPT 3.5 and Gemini: A Comparative Evaluative Study

The study aimed to assess the efficiency of artificial intelligence tools, namely, ChatGPT 3.5 and Gemini in providing reference services in Arab libraries. Through the analysis of their ability to answer a diverse set of inquiries in order to explore the use of such technologies in the Arab libraries. The study employed a descriptive analytical approach and content analysis method. A set of reference inquiries were directed to each tool individually; their accuracy, comprehensiveness, and presentation style were evaluated. The results showed statistically significant differences between ChatGPT 3.5 and Gemini in answering reference inquiries in favor of Gemini, which had more accurate and comprehensive responses. The study proved that Gemini is the preferred option for use in Arab libraries and information institutions, as it has a better understanding of Arabic, accurate responses, comprehensive information, and modern content. The results also showed statistically significant differences between ChatGPT 3.5 and Gemini in answering questions posed by faculty members in the social sciences, humanities, and educational disciplines in favor of Gemini, while no statistically significant differences were found between the two models for questions posed by faculty members in exact and applied sciences.