Acceptance of Artificial Intelligence and Future Italian Physicians’ Perspectives: a cross-sectional descriptive survey

Abstract

Artificial intelligence (AI) is increasingly shaping clinical reasoning and healthcare delivery, yet evidence on physicians’ acceptance, trust, and ethical expectations remains limited. This nationwide cross-sectional study investigated the perceptions of 293 Italian physicians, the majority of whom were hospital-based specialists, regarding the integration of AI into clinical practice. Although 78.5% had previously used AI tools, only 4.8% had received formal training, revealing a significant educational gap. Respondents expressed a cautiously optimistic attitude toward AI, particularly recognizing its potential to enhance diagnostic accuracy and support therapeutic decision-making, while showing moderate agreement on its ability to reduce workload. Persistent reservations were linked to limited trust in autonomous AI outputs and fears of professional displacement. Factor analysis identified three clusters influencing acceptance: primary facilitators such as usability and clinical relevance; secondary enablers including trustworthiness, safety, transparency, and privacy protection; and minor inhibitors related to insufficient expertise, limited peer endorsement, and perceived loss of autonomy. Ethical expectations were notably high, with strong agreement on the need for defined medico-legal responsibilities, independent ethical evaluation, and validation of clinical reliability. Participants also emphasized the importance of equity in AI deployment, regulatory harmonization, and targeted professional education. Overall, Italian physicians appear open to the adoption of AI in medicine, provided that human oversight, fairness, and trust remain central. Bridging the current gaps in training, governance, and ethical regulation is essential to foster a healthcare environment in which AI complements—rather than replaces—human clinical judgement.

Ähnliche Arbeiten