Washington’s latest weapon against healthcare waste isn’t an investigator; it’s an algorithm. The Department of Health and Human Services is rolling out a broad AI-powered audit review system designed to flag financial irregularities across the entire federal health funding network.
A Paper Trail That Was Going Nowhere
For years, thousands of annual compliance reports filed by states, universities, nonprofits, and local governments have piled up with little meaningful follow-up. Under federal rules, any entity spending $1 million or more in federal funds each year must submit an audit — but until now, those documents rarely triggered timely action.
The new program changes that by applying tools like ChatGPT to continuously scan those filings, hunting for patterns that suggest mismanagement or noncompliance. Gustav Chiarello, the HHS official spearheading the effort, put the problem bluntly: “It’s classic big government: Everyone files an audit, and it lands with a thud, and no one does anything about it. Here, with AI, we’re able to dig into it.”
Programs covered include state Medicaid systems, federally funded research, and grant recipients in areas like addiction treatment. Organizations that ignore filing requirements or leave findings unaddressed now risk losing their federal funding.
Critics Aren’t Convinced
Skeptics are raising two distinct concerns: whether AI is reliable enough for high-stakes enforcement, and whether the administration will apply it evenhandedly. Consumer advocacy group Public Citizen’s co-president Rob Weissman was blunt in his skepticism: “The AI is kind of beside the point when you assess what their actual objectives are, rather than what they pretend they are.”
Those doubts aren’t unfounded — the administration previously admitted to using flawed data in at least one Medicaid fraud investigation, and opponents note that enforcement actions have disproportionately targeted Democratic-led states.
Chiarello said the technology is already attracting interest from other parts of the federal government, and he’s actively encouraging adoption: “It would be fairly easy for the other agencies to use our technology and jump on it.”
