USING ARTIFICIAL INTELLIGENCE IN SCIENTIFIC ARTICLE WRITING: A SYSTEMATIC AND BIBLIOMETRIC REVIEW

Abstract

This study investigates the use of artificial intelligence tools in scientific article writing by conducting a systematic and bibliometric review of publications from 2015 to 2025. The research examines how AI technologies particularly large language models such as ChatGPT are integrated into various stages of academic writing, including idea generation, problem formulation, literature analysis, drafting, and language refinement. Using the PRISMA 2020 protocol, 830 articles were initially identified in the Scopus database, of which 50 met the inclusion criteria. Bibliometric analysis was performed using Bibliometrix and VOSviewer to map publication trends, citation dynamics, keyword clusters, and geographical distribution. The results reveal a sharp increase in scholarly interest after 2022, reflecting the widespread adoption of generative AI tools. The cluster analysis identifies several major research directions, including the effectiveness of AI in academic writing, issues of ethical integrity, technological development of LLMs, and the pedagogical implications of AI use in higher education. Despite the documented benefits such as improved productivity, clarity, and accessibility significant concerns remain regarding hallucinations, fabricated references, academic honesty, and authorship attribution. Overall, the review demonstrates that AI has become a transformative yet contested tool in scientific writing and highlights the need for responsible, transparent, and critically informed use of AI-assisted technologies in academic research