> ## Content Index
> Fetch the complete content index at: https://www.fdaweb.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# Post Questions FDA Role in AI/ML Oversight
- URL: https://www.fdaweb.com/post-questions-fda-role-in-ai-ml-oversight/
- Published: 2019-08-14T12:00:00.000Z
- Updated: 2026-09-15T01:43:43.000Z
- Author: David McFarland
- Tags: Devices, #legacy-id-D5144770

Forbes Technology Council member **Charles Aunger** is questioning whether FDA is the appropriate agency to exercise regulatory authority over artificial intelligence/machine learning (AI/ML)-based software as a medical device. Writing in a *Forbes* online [post](https://www.forbes.com/sites/forbestechcouncil/2019/08/14/should-the-fda-regulate-al/?ref=fdaweb.com#3ec9bb1039a0), Aunger says FDA has released a proposed framework and stated its intention to “apply our current authorities in new ways to keep up with the rapid pace of innovation.”

“Proposing a regulatory framework for medical AI/ML software modification via a ‘predetermined change control plan’ is cause for concern,” he writes. “And should FDA be regulating such technology at all?”

The post acknowledges that AI requires oversight, particularly in something as critical as a medical application. “Creating guardrails for AI/ML usage is urgent and necessary,” Aunger says. “At issue is how those guardrails are developed, applied, managed, and by whom.”

The post reviews the varied applications of AI/ML technologies and the oversight being considered by the federal Transportation Department and Federal Aviation Administration in addition to FDA. “Are all these agencies and entities grappling with AI/ML even talking to each other?” he asks.

Aunger cautions that the healthcare sector does not have the best track record of implementing any kind of new process, infrastructure, or technology. “Enormous bureaucracies are involved,” he says. “The healthcare industry moves slowly when lives are at stake. Regulatory bodies like FDA move slowly since resources are scarce. But technology moves very quickly, and AI/ML technology moves more quickly than anything we’ve seen before.”

Aunger says he does not think it is possible to apply “current authorities in new ways” and hope to keep pace. The pace of AI/ML is just too fast, he says, and the opportunities for both catastrophic failure and awe-inspiring advancement are too great. He suggests that what may be needed is a new kind of overarching AI/ML cyber command that is actively and continuously addressing the complicated AI/ML evolutionary issues at a more fundamental level. “It could be accessed and implemented vertically by industry sector and area of expertise,” he says, “and would require a lot of input and engineering and collaboration, and commitment from people with conflicting priorities who don’t usually cooperate. But it could provide a feasible mechanism for managing adaptive technology more strategically than at present. The question then becomes, is FDA the right place to do that?”