Artificial Intelligence in Exercise Prescription in Palliative Care: Perceptions and Ethical Issues.

Background/Objectives: The integration of artificial intelligence (AI) in healthcare has been progressively expanding, with growing interest in its potential application in palliative care, particularly in exercise prescription. However, there is limited scientific evidence addressing professionals' perceptions of this use. This study aimed to explore Portuguese healthcare professionals' perceptions of AI in exercise prescription for palliative care and to identify related ethical implications and training needs. Methods: A cross-sectional, descriptive, and analytical study was conducted using an online questionnaire applied to health professionals. Sociodemographic data, knowledge, and perceptions of AI in exercise prescription in palliative care were collected. Descriptive and inferential statistical analyses were performed, including Mann-Whitney and Spearman correlation tests. Results: The sample consisted mainly of young female professionals with backgrounds in physiotherapy. Most participants reported little knowledge and experience with AI in clinical practice but expressed a strong interest in learning. Perceptions regarding the usefulness of AI were neutral to slightly positive, particularly concerning quality of life and personalization of interventions. Ethical aspects were viewed positively, especially the complementarity of AI to human clinical judgement, transparency, and explicability. Spearman correlations indicated that greater AI literacy and longer clinical experience were associated with more positive perceptions. Conclusions: The findings highlight the importance of developing structured training programmes that integrate practical, ethical, and technical components for the safe and responsible use of AI in palliative care. Ethical guidelines are proposed to ensure the humanization of care and the preservation of patient autonomy when implementing AI in exercise prescription.