Health systems urge government action to support AI transparency

Health care leaders are navigating the complex challenge of creating transparent AI governance while managing potential risks in sharing sensitive implementation data. At a recent Newsweek webinar, experts from the Coalition for Health AI (CHAI), legal practice, and healthcare institutions discussed the tensions between building collaborative knowledge about health AI performance and protecting organizations from liability. Their discussions highlighted how health AI’s rapid evolution requires new frameworks for sharing outcomes data while providing necessary legal protections for participating organizations—a balance that may ultimately require government intervention to create appropriate incentives for transparency.

The big picture: CHAI is developing a public registry for health AI model cards that would function like “nutrition labels” for AI tools and create an industry-wide database of implementation outcomes.

• The registry aims to serve as a “post-market or post-deployment monitoring network” where organizations can track how AI models perform across different populations and geographies.
• This centralized approach would help identify both degrading model performance and unexpected positive outcomes, contributing to the collective understanding of health AI applications.

Key challenges: Healthcare organizations face significant disincentives to sharing their AI implementation data despite the potential collective benefits.

• Duke Health’s Dr. Michael Pencina acknowledged that while his institution is a CHAI founding member, privacy concerns and organizational reluctance remain barriers to full participation.
• Legal expert Danny Tobey noted that in the current unregulated market, organizations take on risk by participating in transparency initiatives.

Why this matters: Transparency about AI performance across different healthcare settings is crucial as new AI capabilities emerge weekly with unknown consequences.

Potential solutions: Experts suggested that government intervention may be necessary to create the right balance of incentives and protections.

• Tobey proposed that legislators could create “safe harbor or presumptions of prudence” for organizations participating in voluntary disclosures.
• CHAI is exploring an AI-specific patient safety outcomes registry to enhance protections for participating members.

What they’re saying: Health AI leaders emphasized both the value and risks of transparency in implementation data.

• “What we’re trying to do in CHAI is create a public space where health systems can safely share that information, in terms of the best practices, of how to deploy and how to use [AI],” explained Dr. Brian Anderson, CHAI’s co-founder and CEO.
• Dr. Andreea Bodnari of Alignmt.AI highlighted the actionable benefits: “If you’ve deployed an ambient AI tool and you receive a notification that there’s a vulnerability on a specific patient population, that’s something you have to act on right away.”