The American Medical Association (AMA) Center for Digital Health and AI is advancing a framework to clarify liability and accountability when artificial intelligence (AI) systems used in clinical care cause patient harm. The proposed approach aims to align responsibility with each party's authority and ability to prevent or mitigate harm.
Existing liability frameworks were not designed to address clinical AI systems, where physicians, technology developers, and healthcare organizations have different levels of control over how these technologies are developed, implemented, and used. This can expose physicians to liability for risks beyond their reasonable control while making it difficult for patients to obtain appropriate remedies when AI-related failures result in harm.
Under the proposed framework, developers and manufacturers would be responsible for demonstrating and maintaining the safety and effectiveness of their products and addressing risks within their control. Healthcare organizations and other entities deploying clinical AI systems would be accountable for their selection, implementation, governance, and monitoring in real-world settings.
Physicians would be responsible for the clinical judgments they can reasonably exercise, taking into consideration the system's intended use, known limitations, and information available at the point of care. Patients would also retain meaningful avenues to seek accountability and appropriate remedies when harm results from defective AI systems or other actionable failures.
The framework recognizes that liability may be shared among multiple parties when clinical AI systems cause harm, with responsibility determined by each party's ability to prevent or mitigate the risks. It also clarifies that physician oversight alone cannot replace adequate evidence of safety and effectiveness, appropriate system design, or the conditions required for meaningful clinical review.
Click here to read the original press release.
More News in this Theme
