AI in health care decisions the subject of Georgia bill

By Susan Rupe

Although Georgia State Rep. Mandisha Thomas, D-District 65, said she believes artificial intelligence is “a wonderful tool,” she does not believe AI should have the final word when it comes to health care decisions.

Thomas is the sponsor of House Bill 887, which outlines restrictions on the use of artificial intelligence in insurance coverage, health care and public assistance. The bill would prohibit health care decisions based “solely on results derived from the use or application of artificial intelligence or utilizing decision tools.”

Georgia is one of the first states in the U.S. to introduce a bill addressing the use of AI in health care and insurance. Thomas told InsuranceNewsNet the bill defines “health care” as “treatment procedures or services from any health care clinician, whether it's prescribing a medication, or whether it's creating a care plan for that patient, on down to any nutritional plans for patient - just any services or procedures that are provided.”

The bill requires that AI-generated health care decisions be reviewed by an individual with “authority to override” the tools’ existing decision. It also requires the Georgia Composite Medical Board to establish policies — including, but not limited to, disciplining physicians. In essence, the bill prohibits physicians from relying solely on AI to make clinical decisions. The bill proposes a similar approach regarding AI and automated decision-making tools in insurance coverage and public assistance.

“We’re not against AI,” Thomas said. “We're just saying that AI cannot have the last word, that it can't replace an employee or a skilled worker. AI can be part of the process but the last word has to be overseen by a skilled worker.”

Thomas said her bill was prompted by a constituent’s experience with her physician using AI to recommend a prescription drug. The doctor reviewed the AI recommendation and found that the drug was suitable, but the AI-recommended dosage was much too high and could possibly kill the patient.

“Health care theories and standard practices don’t go away just because you have a new tool in AI,” Thomas said. “We need to include AI but we need to be sure it makes sense for patients and their needs.”

Susan Rupe is managing editor for InsuranceNewsNet. She formerly served as communications director for an insurance agents' association and was an award-winning newspaper reporter and editor. Contact her at [email protected]. Follow her on X @INNsusan.