What You Should Know:
– The Centers for Medicare & Medicaid (CMS) has launched a new market-based research challenge called the Crushing Fraud Chili Cook-Off Competition. The goal of the competition is to use explainable artificial intelligence (AI) and machine learning (ML) models to identify anomalies in Medicare Fee-for-Service (FFS) claims data that can be translated into new indicators of fraud.
– The challenge also seeks scalable technologies that can reduce labor-intensive processes while keeping humans involved in the loop for effective oversight and interpretability.
Medicare Impact of Americans
Medicare is a lifeline for millions of Americans, but its size and complexity make it vulnerable to fraud, such as false billing and upcoding. Fraudulent activity diverts critical resources away from patient care and undermines public trust. By using innovative, data-driven approaches like explainable AI/ML, CMS aims to proactively identify suspicious behavior and take timely, informed action to protect public resources.
Why Explainable AI is Critical
CMS is specifically seeking explainable AI/ML techniques because pattern detection alone is not enough to prove fraudulent behavior in legal or enforcement contexts. Explainable AI/ML makes insights transparent and accessible to program integrity teams, regulators, and policymakers, which in turn builds trust in the system and enables fair and accountable enforcement actions.
To maximize impact, solutions must be scalable and focus on uncovering broader patterns and systemic vulnerabilities that affect large groups of claims or entities. This approach allows for a more efficient allocation of enforcement resources and a greater return on oversight efforts.
Competition Details
The competition will take place in two phases:
- Phase 1 – Proposed Technology Development: CMS is inviting research proposals from all interested parties. The proposals must be tailored to the Medicare FFS claims data that will be provided in Phase 2. Ten teams will be selected as finalists to advance to the next phase. The submission period for Phase 1 is from August 19 to September 19, 2025.
- Phase 2 – Competition: Finalists will receive access to the 2022-2024 Standard Analytical Files (SAF) Limited Data Sets (LDS) containing claims data for a random 5% sample of Medicare beneficiaries. Participants will then apply their proposed explainable AI/ML techniques to the data and submit a summary of their findings. Phase 2 submissions are due by December 1, 2025.
The CMS Crushing Fraud Chili Cook-Off Competition is open to all U.S.-based teams, academic institutions, nonprofit organizations, and private sector entities. The winning team will be publicly announced by CMS via its social media channels the week of December 15, 2025.
