Consequences, Trends, and Future Directions in Using AI in Science Education: Is it a Boon or a Bane?

Abstract The study aims to examine the consequences, trends, and future directions of using artificial intelligence (AI) in science education, addressing the central question of whether AI serves as a boon or a bane for teaching and learning. It seeks to map the global research landscape, identify benefits and challenges, and provide evidence-based insights to guide responsible adoption. A bibliometric analysis of 1,437 Scopus-indexed publications from 1990 to 2025 was conducted. Data were analyzed using VOSviewer for co-authorship, keyword co-occurrence, and network visualization, and Bibliometrix (RStudio) for descriptive statistics, publication trends, thematic mapping, and citation patterns. The analysis covered authorship, collaboration, institutional and country contributions, thematic clusters, and research evolution. The study reveals exponential growth in AI and science education research, with an annual growth rate of 18.02%. The United States, China, and Germany emerged as leading contributors and hubs of collaboration. Thematic mapping identified four dominant domains: (1) technological innovations (machine learning, generative AI, large language models), (2) pedagogical integration (adaptive tutoring, inquiry learning, curriculum reform), (3) disciplinary applications across STEM fields, and (4) interdisciplinary extensions into health and social sciences. Despite AI’s promise to enhance engagement, personalization, and innovation, gaps remain in addressing ethics, teacher preparedness, equity, and participation from developing regions.