Indiana tests AI to flag Medicaid fraud

Indiana’s Family and Social Services Administration is joining CMS and Oracle in a 90-day pilot to test whether AI can detect suspicious Medicaid billing, prevent improper payments and generate case packages for investigators. FSSA Deputy Chief of Staff Marcus Barlow said the goal is to ensure Medicaid dollars are spent correctly and to make the programs more sustainable.

The system is expected to review claims in real time, flagging potential waste, fraud and abuse before payments are made. Barlow said the tools will examine Medicaid providers as well as the more than 1 in 4 Hoosiers who use some form of Medicaid. A data use agreement says Indiana will remove personal and protected health information before sharing data with Oracle, which must destroy restricted information it receives and cannot share FSSA data with outside parties.

Medicaid spending has been a major focus for Gov. Mike Braun’s administration. FSSA says Indiana’s Medicaid appropriations have jumped 134%, from roughly $2.1 billion in 2017 to nearly $5 billion for fiscal year ‘27. CMS will review a proof-of-concept report after the pilot to assess challenges and possible expansion to other states.