Algorithmic moderation of medical knowledge as an epistemic challenge of the digital transformation of medicine

Abstract

The research examines the epistemic challenges arising from the increasing role of algorithmic moderation in the digital transformation of medicine. It explores how digital platforms, search engines, and artificial intelligence technologies, particularly large language models (LLMs), reshape the production, evaluation, validation, dissemination, and practical use of medical knowledge. Special attention is devoted to the epistemic risks associated with algorithmic mediation, including algorithmic opacity, the reproduction of systematic biases, informational homogenization, hallucinations generated by large language models, and the diminishing role of contextual clinical judgment in medical decision-making. The paper also traces the historical development of medical knowledge standardization and demonstrates how AI-driven technologies redefine traditional mechanisms of knowledge production and professional expertise. Five conceptual approaches to understanding the role of artificial intelligence in medicine are identified and systematized: the algorithm-centered approach, the critique of digital paternalism, the synergistic approach, the institutional-pragmatic approach, and the institutional-sociotechnical approach. These approaches reflect different conceptions of the relationship between physicians, algorithmic systems, and healthcare institutions, as well as different models for allocating epistemic authority and responsibility. The paper concludes that the future of digital medicine depends not on replacing physicians with artificial intelligence but on developing transparent, accountable, and human-centered models of human–AI collaboration that strengthen clinical decision-making while preserving professional responsibility and epistemic autonomy.