CDRH Director Tarver Tells AAMI neXus Final AI Lifecycle Guidance Is Coming

The Food and Drug Administration's device center plans to issue final guidance on managing artificial intelligence-enabled devices across their life cycle, Center for Devices and Radiological Health Director Michelle Tarver told an industry standards conference, MedTech Dive reported on April 22.

Tarver gave an update on digital health and AI policy on April 17 at the Association for the Advancement of Medical Instrumentation's neXus conference. The FDA is reviewing comments on the draft to produce the final version, she said, according to the report. She did not give a date.

The draft on the table

The FDA issued the draft in January 2025. In its announcement at the time, the agency said the guidance, if finalized, "would be the first guidance to provide comprehensive recommendations for AI-enabled devices throughout the total product lifecycle." It covers how and when sponsors should describe postmarket performance monitoring in marketing submissions, and sets out the agency's current thinking on strategies to address transparency and bias, the FDA said.

At neXus, Tarver said all AI algorithms are required to be trained on data that reflects the intended use population and should be validated in that population, MedTech Dive reported. She also addressed postmarket monitoring once devices are in use, noting that the technology can hallucinate, that bias can come into play and that models may drift and become less accurate over time.

"All of that can impact the output that those technologies are generating and impact patient outcomes as a result," Tarver said, according to the report.

Generative AI next

Tarver also pointed to generative AI, which can produce text, images, software code and other outputs. CDRH's Digital Health Center of Excellence held two advisory committee meetings on the topic: one in 2024 on how the agency should regulate generative AI, and another last year on digital mental health devices, including chatbots, MedTech Dive reported.

"We are taking all the feedback that we heard, and I encourage you to look toward the end of this year for some initial thoughts," Tarver said.

The remarks come as the center has issued little final AI guidance since the start of President Donald Trump's second term. MedTech Dive noted that in 2025 CDRH issued just one final guidance related to AI, on predetermined change control plans, while the agency dealt with staff cuts that affected experts in AI and digital health, some of whom were later reinstated. This year began with two unannounced changes to digital health guidance documents that exempted some wellness features and clinical decision support tools from device regulation, the publication reported.

The FDA said in its January 2025 announcement that it had authorized more than 1,000 AI-enabled devices through established premarket pathways.