[Proposal for Responsible Use of Generative Artificial Intelligence in Medical Practice].

Abstract

Introduction: The advancement of artificial intelligence (AI), particularly generative AI, has significantly transformed the field of medicine, impacting healthcare delivery, medical education, and research. While the opportunities are substantial, the implementation of AI also raises important ethical and technical challenges, including risks related to data bias, the potential erosion of clinical skills, and concerns about information privacy.

Development: AI has demonstrated great potential in optimizing both clinical and educational processes. However, its operation based on probabilistic prediction is inherently prone to errors and biases. Healthcare professionals must be aware of these limitations and advocate for a transparent, responsible, and safe integration of AI, while maintaining full ethical and legal responsibility for clinical decisions. It is essential to safeguard traditional clinical competencies and prioritize the use of AI in automating low-value, repetitive tasks. In biomedical research, transparency and independent validation are crucial to ensure the reproducibility of findings. Similarly, in medical education, structured training in AI is vital to enable professionals to apply these tools safely and effectively in clinical practice.

Conclusions: Generative AI offers a transformative potential for medicine, but its adoption must be guided by rigorous ethical standards. Comprehensive training, risk mitigation, and the preservation of core clinical skills are essential pillars for its responsible implementation. This transformation must be led by the medical profession to ensure a patient-centered approach to care.