Key points are not available for this paper at this time.
• Artificial intelligence facilitates the shift from hypothesis-driven to data-driven research by improving the discovery and synthesis stages.. • Tools such as ChatGPT and Scite.ai simplify literature review, hypothesis generation and citation analysis. • Data quality is critical: poor data can lead to opaque models, impacting reproducibility and reliability. • Explainable AI (XAI) frameworks are emerging to address issues of transparency and interpretability in research. • Human expertise remains essential for validation and to ensure ethically valid and reproducible AI results. This work offers a critical and evidence-based synthesis of the conceptual, methodological, and social implications of artificial intelligence (AI) in scientific research, significantly enriched by an informetric perspective. The analysis transcends descriptive overviews and simple cataloging of products, providing a deeper understanding of the opportunities AI presents, such as accelerated data analysis, hypothesis generation, and drug discovery. At the same time, crucial challenges that AI introduces are explored, including knowledge monocultures, algorithmic bias, reproducibility issues, and the impact on research integrity and evaluation. The original contribution of this paper lies in the integration of informetric analysis to quantify the influence of AI on the production and dissemination of scientific knowledge, highlighting both its potential as an analytical tool and the risk of bias in the academic record. The paper emphasizes the need for frameworks that harmonize technological capabilities with the irreplaceable ingenuity of human thought, promoting balanced collaboration between AI and researchers, where AI serves as a tool to increase productivity and human oversight ensures ethical rigor, critical evaluation, and creative exploration.
Branda et al. (Sun,) studied this question.