Liability for Use of Artificial Intelligence in Medicine

While artificial intelligence has substantial potential to improve medical practice, errors will certainly occur, sometimes resulting in injury. Who will be liable? Questions of liability for AI-related injury raise not only immediate concerns for potentially liable parties, but also broader systemic questions about how AI will be developed and adopted. The landscape of liability is complex, involving healthcare providers and institutions and the developers of AI systems. The landscape is also evolving, as the technology remains new and explicit legal consideration is relatively scarce. In this chapter, we consider these three principal loci of liability: individual healthcare providers, focused on physicians; institutions, focused on hospitals; and developers. We describe current doctrine, make tentative predictions and outline potential systemic implications.