<p>The integration of Artificial Intelligence (AI) in health systems has the potential to revolutionize healthcare delivery, but it also raises significant concerns regarding bias, transparency, and accountability. This paper proposes a novel “Do Good” framework that prioritizes patient-centered design, transparency, and accountability in AI development and deployment, addressing the limitations of existing approaches. By combining human rights, bioethics, and health systems principles, our framework provides a comprehensive and proactive approach to ensuring that AI systems are safe, effective, and equitable. Our analysis highlights the need for a paradigm shift in healthcare ethics and practice, from a traditional “Do No Harm” principle to a more proactive “Do Good” approach, enabled by AI. We demonstrate the applicability of our framework through a critical examination of existing regulatory processes, health technology assessment, and public procurement of AI, and identify key areas for future research and development.</p>