Ethics of Artificial Intelligence in Respiratory Medicine: A Narrative Review

Artificial Intelligence (AI) has become a transformative tool in the armamentum of respiratory medicine, offering applications in diagnostics, risk stratification, and clinical decision support. Leveraging machine learning (ML) and deep learning (DL), AI systems analyse large datasets— from imaging to electronic health records — to assist clinicians in managing respiratory conditions. In diseases such as COPD, interstitial lung disease (ILD), lung cancer, and obstructive sleep apnea (OSA), AI has achieved diagnostic accuracy comparable to human experts. ML models predict COPD exacerbations and OSA severity, while convolutional neural networks (CNNs) enhance lung nodule detection and ILD classification. Despite promising results, most systems lack prospective validation and regulatory approval. The use of AI also raises critical ethical issues, including bias and fairness, algorithmic transparency, patient autonomy, data privacy, and liability in decision-making. To ensure trustworthy integration of AI into respiratory care, future development must be patient-centered, ethically grounded, and supported by robust regulatory frameworks.