Artificial intelligence in inquiry-based science teaching: A bibliometric study

The integration of artificial intelligence (AI) into science education has accelerated in recent years, raising urgent questions about its role in supporting inquiry-based pedagogies. This study employs a bibliometric methodology to map global research trends, subject areas, geographical distribution and conceptual structures at the intersection of AI and inquiry-based learning. A Scopus query retrieved 26,283 documents, which were analyzed through descriptive statistics and science mapping. Findings reveal a sharp rise in publications since 2017, with notable peaks in 2024 and 2025, reflecting the rapid mainstreaming of AI in science education. Subject area analysis indicates that while research is anchored in the social sciences and computer science, it has progressively diversified to encompass psychology, mathematics, engineering and medicine. Geographically, the USA, China, and the UK dominate output, but growing contributions from Indonesia, India and Malaysia highlight increasing global engagement. Keyword co-occurrence analysis identified five thematic clusters: technological foundations, pedagogical applications, human and social dimensions, learner engagement, and classroom practice. By providing a large-scale bibliometric analysis of research at the intersection of AI and inquiry-based science education, this study establishes a foundation for future scholarship while underscoring unresolved challenges surrounding equity, ethics and collaboration. The findings also provide evidence to inform policy initiatives, teacher training programs and equitable investment in AI-enhanced science education.