The digital transformation in education has brought about various technology-based learning innovations, including the use of artificial intelligence and search engines. This study aims to analyze the differences in Civics (PKn) learning outcomes between students using ChatGPT and those using Google as a learning medium. This study used a quasi-experimental approach with a nonequivalent control group design. The subjects consisted of 60 fourth-grade students divided into two groups: the ChatGPT group (30 students) and the Google group (30 students). The research instrument was a learning outcome test consisting of 10 multiple-choice questions and 4 essay questions. Data analysis was conducted through descriptive and inferential tests using the Mann–Whitney test. The results showed a significant difference between the learning outcomes of students using ChatGPT and those using Google. The ChatGPT group showed a higher improvement in scores than the Google group. This finding indicates that dialogic AI-based learning is more effective in improving elementary school students' understanding of Civics concepts.
Copyrights © 2026