Position: Specialty Society-Led Meta-Governance is Essential to Responsible Implementation of Generative AI in Cardiovascular Care

Yumin Gao, Jacqueline Sandling, Francis Arellano, Zahra Ahmad, Rohan M. Rani, Deeya Garg, Emily Y Chen, Aamir Javaid, Francoise Marvel, Seth Martin, Raj Ratwani, Charles German

Keywords: Generative artificial intelligence, Large language models, Cardiovascular care, Governance

TL;DR: Generative AI in cardiology shows promise but requires validation standards, monitoring infrastructure, reimbursement guidance, and specialty society meta-governance to guide responsible implementation and ensure safe, equitable care.

Abstract: Generative artificial intelligence (AI), particularly large language models (LLMs), is rapidly emerging in cardiology, the leading global cause of death and a major determinant of population health. While the development of AI applications in cardiology has expanded steadily and now ranks second only to radiology in FDA-approved AI-enabled devices, the use of generative AI models in real-world clinical settings remains very limited. Current efforts focus primarily on exploratory studies related to documentation and patient education rather than direct clinical decision support and disease management. Four barriers define the current landscape: 1) insufficient frameworks for external and local validation, 2) sensitivity to contextual and user factors, 3) lack of structured post-deployment governance, and 4) misaligned reimbursement and regulatory incentives. Addressing these challenges will require coordination between regulators and specialty societies such as the American Heart Association (AHA) and American College of Cardiology (ACC). Society-led meta-governance anchored in validation standards, monitoring infrastructure, and reimbursement guidance is essential to ensure the safe, equitable, and effective implementation of generative AI in cardiovascular care.

Submission Number: 103