AI-Powered Claims Intelligence for Identifying Billing Anomalies and Fraud in Medicare and Medicaid

Abstract

Medicare and Medicaid fraud account for more than $60 billion in annual losses in the United States, placing a significant burden on taxpayers and eroding the integrity of the healthcare system. Traditional rule-based systems struggle to keep up with evolving fraud patterns and complex billing behaviors. This article presents an unsupervised machine learning (ML) approach designed to autonomously identify abnormal billing activities by detecting outlier trends and CPT code mismatches across provider datasets. Using real-world data from outpatient facilities, the model demonstrated a 52% increase in fraud detection accuracy and reduced manual audits by 40%, significantly enhancing operational efficiency. By integrating AI into the audit lifecycle, healthcare agencies and insurers can proactively target high-risk providers, ensure compliance, and optimize resource allocation.This research advocates for the adoption of regulatory tech solutions and intelligent audit pipelines that adapt to changing fraud patterns, ultimately contributing to a more transparent and accountable public health ecosystem.