This study is motivated by the rapid development of Artificial Intelligence (AI) in education, particularly its use as a learning resource in Islamic Religious Education (PAI). One of the widely used AI technologies is Gemini, which provides instant answers to various questions. However, concerns arise regarding the validity and accuracy of its responses when compared to authoritative Islamic sources. Therefore, this research aims to analyze the validity of Gemini AI’s answers on the topic of wudhu in PAI by comparing them with the classical fiqh book Fathul Qarib. This study employs a qualitative approach with a library research method. Data were collected through structured prompting techniques to obtain AI-generated responses related to wudhu, which were then analyzed using content analysis and comparative descriptive methods. The findings reveal that most of Gemini AI’s answers are highly consistent with the explanations found in Fathul Qarib, particularly in terms of the fundamental aspects of wudhu such as its pillars, conditions, and nullifiers. Nevertheless, some responses tend to provide broader explanations by incorporating additional perspectives beyond the scope of the book. Overall, Gemini AI demonstrates a high level of validity and can be considered a supportive learning resource in PAI, although verification with authoritative fiqh references remains necessary to avoid misconceptions in religious understanding.
Copyrights © 2026