Toward an Ethical Framework for Patient-Facing Artificial Intelligence in Clinical Settings

Area of Science:

• Medical Ethics
• Health Informatics
• Artificial Intelligence in Healthcare

Background:

• Patient-facing artificial intelligence (AI) is increasingly integrated into clinical settings, shifting patients from passive recipients to active participants in AI-augmented healthcare.
• This transition necessitates a thorough examination of the ethical considerations surrounding the implementation of AI tools that directly interact with patients.

Purpose of the Study:

• To identify and analyze the ethical considerations associated with implementing patient-facing AI in clinical settings.
• To develop a preliminary ethical framework for guiding the responsible deployment of AI in healthcare, using a postpartum depression prediction tool as a specific use case.

Main Methods:

• Conducted 36 qualitative interviews with diverse stakeholders: patients, healthcare professionals, and AI developers.
• Utilized a postpartum depression prediction tool as a concrete use case to explore ethical challenges in patient-facing AI.

Main Results:

• Identified six interconnected themes that formed a preliminary ethical framework.
• The framework highlights transparency as fundamental, balances core ethical principles (autonomy, beneficence, justice, privacy), and recognizes trust as crucial for engagement.
• Ethical implementation involves acknowledging and navigating trade-offs between principles, rather than a mere compliance approach.

Conclusions:

• Ethical implementation of patient-facing AI requires explicit attention to the interplay between core principles and the inherent trade-offs.
• Engaging patients in deliberating context-specific ethical tensions is vital for successful and responsible AI adoption in healthcare.
• Trust emerges as both a prerequisite for and an outcome of ethical patient-AI engagement.