Algorithmic Oracles and Humanistic Dilemmas: Philosophical Reflections on Large Language Models in Medical Artificial Intelligence

Abstract

Medical artificial intelligence demonstrates immense potential in optimizing clinical workflows and democratizing access to high-quality healthcare resources. However, large medical language models are deconstructing and reconstructing traditional medical authority, raising complex philosophical and ethical questions. This paper examines how these models challenge and reshape medical authority, revealing uncertainties regarding responsibility attribution, professional identity, and power structures. It analyses their impact on doctor–patient relationships, highlighting how empathy deficits, shifting trust, and algorithmic bias may lead to the "dehumanization" of medical practice. Furthermore, it notes that when algorithms intervene in end-of-life care and decisions concerning the value of life, their quantitative rationality fails to address core concerns of dignity, meaning, and compassion. Reaffirming medicine's humanistic core and establishing physicians as the ultimate responsible agents, this study proposes a novel human–machine collaborative ethical framework centered on patient welfare.

Ähnliche Arbeiten