Experts suggest that generative AI models in medicine should be licensed similarly to doctors and nurses to ensure accountability and safety.
Key Details
- 1Academic, policy, and public health leaders propose a licensure framework for medical GenAI models, similar to those for clinicians.
- 2UPenn's Eric Bressman and collaborators published a JAMA Internal Medicine opinion highlighting this approach.
- 3A blog post by UPenn policy expert Julia Hinkley supports a federal digital licensing board overseeing medical AI.
- 4FDA could handle premarket review, while state boards might provide ongoing oversight and coordinate standards.
- 5The initiative aims to address concerns about AI hallucinations, performance drift, and maintain standards and accountability.
Why It Matters
As generative AI’s use grows in radiology and other areas, robust and harmonized regulation is critical to maintain quality, safety, and trust. Introducing a licensing system modeled after medical professions could help set clear expectations and mechanisms for oversight of clinical AI systems.
Source
HealthExec
Related News
•AuntMinnie
ONC Official at SIIM Calls for Smarter AI and Better Image Sharing
ONC's Dr. Jason Funderburk urges improvements in medical image sharing and AI readiness at SIIM 2024.
•HealthExec
House Moves to Block AI-Driven Medicare Prior Authorization Pilot
A U.S. House committee voted to block the expansion of an AI-driven Medicare prior authorization pilot active in six states.
•Radiology Business
Navigating Payment and Reimbursement for Radiology AI Solutions
Radiology AI reimbursement in the US remains complex and fragmented, complicating provider and vendor payment strategies.