The news: Republican lawmakers are opposing plans to regulate the use of AI in healthcare—in particular a Department of Health and Human Services (HHS) proposal to create assurance labs that would vet AI tools, according to Politico.
Assurance labs are part of a framework established by the Coalition for Health AI (CHAI), a group that includes the Mayo Clinic, Microsoft, and Johns Hopkins. These labs are designed to assess AI technologies using agreed-upon testing standards.
What are lawmakers arguing? Reps. Dan Crenshaw (R., Tx.), Brett Guthrie (R., Ky.), Jay Obernolte (R., Ca.), and Mariannette Miller-Meeks (R., Ia.) have asked the HHS to reconsider its support of CHAI’s framework, arguing that government-led labs could foster conflicts of interest and erode the public’s trust in both the government and tech companies.
Yes, but: The lawmakers’ position stands in stark contrast to physicians’ beliefs concerning the responsible use of AI in healthcare.
89% of US physicians need to know healthcare AI tech is properly sourced and vetted before they’ll even use it, per Wolters Kluwer Health. Some health systems have been hesitant to incorporate AI solutions into doctors’ clinical workflows due to a lack of agreed-upon industry standards. And 91% of physicians want to be involved when adopting AI tools for their practice, per the AMA.
Our take: Conservatives have historically championed deregulation and lawmakers’ recent AI-centered efforts are an extension of that practice.
With a Republican administration set to take office in 2025, we expect a methodical undoing of President Biden’s AI oversight policies to take place. We’ll be keeping our eyes on whether this unraveling discourages physicians from using healthcare AI tech, considering lagging regulations and guidance.