Recently, the U.S. Food and Drug Administration (FDA) released a discussion paper regarding factors that go into the regulation of generative AI–enabled medical devices. The regulatory office is seeking public feedback on the paper relating to the regulation of generative AI–enabled medical devices, focusing on areas of risk assessment, premarket evaluation, and postmarket monitoring.

“Artificial intelligence is transforming medicine, and the United States must lead in shaping how this technology is developed and used safely and responsibly,” stated Kyle Diamantas, JD, Acting FDA Commissioner. “[This] announcement reflects the FDA’s commitment to advancing innovation for health-care professionals and leveraging AI to improve care and patient health outcomes.”

The discussion paper reflects the FDA's efforts to follow the Trump Administration's priorities to harness AI to accelerate the timeline for innovative medical products to go from bench to bedside; the effort also aligns with the FDA's Public Health Pillar on Innovation and Global Leadership. "Considerations for the Regulation of Generative AI-Enabled Medical Devices" is led by the Digital Health Center of Excellence within the FDA's Center for Devices and Radiological Health.

“Generative AI–enabled medical devices are poised to reshape the health technology landscape, and the FDA has an important responsibility to provide thoughtful leadership for this new era,” said Rick Abramson, MD, Director of the FDA Digital Health Center of Excellence. “This discussion paper advances the frontiers of regulatory science and propels a critical conversation about how to enable beneficial innovation, protect public health, and preserve trust.”

“Patients and clinicians deserve a regulatory approach that keeps pace with the rapid innovation of digital health technologies,” said Michelle Tarver, MD, PhD, Director of the FDA Center for Devices and Radiological Health. “By inviting input from the public, we are launching a transparent process to inform the development of an approach that safeguards patients and consumers, advances innovation, and serves as a potential model for regulators around the world.”

The discussion paper lays out the unique nature of generative AI–enabled medical devices compared with usual FDA regulated medical devices, due to their vast potential, wide-ranging characteristics, continuous adjustments and learning, and potential risks. A possible two-axis framework is laid out for risk assessment to inform regulatory expectations as well as possible approaches to premarket evaluation and risk-proportionate postmarket monitoring. The framework considers the increased risk of a device according to the independence of device activity as well as the consequences of relying on an incorrect output. Such a framework provides a simple way of thinking about risk collectively across the range of generative AI–enabled medical device functions, from informational non-directive actions to fully autonomous clinical AI.

In the discussion paper, the Center for Devices and Radiological Health acknowledges that existing performance evaluation approaches may be inadequate for generative AI–enabled medical devices and their function, and so new methodologies are needed. The agency suggested a competency-based approach for premarket evaluation, including both non-clinical device benchmarking—focused on elements of safety, clinical proficiency, generalizability, and agentic AI competencies—and clinical confirmation—through retrospective evaluation, prospective clinical studies, clinician adjudication, standardized patient interactions, or shadow deployment.

Specific discussion questions are introduced throughout the paper, designed to guide feedback and inform the development of the resulting regulatory framework.

The discussion paper is intended to spark discussion among stakeholders about a regulatory framework for generative AI–enabled medical devices, but it is not a draft or final guidance. Public feedback on the FDA discussion paper is requested by October 19, 2026.

ASCO AI in Oncology is published by Conexiant under a license arrangement with the American Society of Clinical Oncology, Inc. (ASCO^®). The ideas and opinions expressed in ASCO AI in Oncology do not necessarily reflect those of Conexiant or ASCO. For more information, see Policies.