Ensemble and Cohere Build First RCM-Native AI Model for Healthcare Revenue Cycle

The model is intended to streamline complex workflows from patient intake to account resolution while improving accuracy and reducing revenue leakage.

Ensemble, a 12-year-old revenue cycle management (RCM) provider, has teamed up with enterprise AI company Cohere to develop the first RCM-native large language model aimed at healthcare financial operations. The model is intended to streamline complex workflows from patient intake to account resolution while improving accuracy and reducing revenue leakage.

The RCM-native LLM is trained using Ensemble’s operational expertise, documented procedures, industry patterns, payer trends, and denial behaviors, supported by synthetic, deidentified datasets in a HIPAA-compliant environment. Executives stressed that no identifiable patient data or PHI is used for training.

"RCM is deeply procedural and conditional. We wanted to go beyond context engineering and retrieval-augmented generation to build a truly trained model for the industry," said Ensemble CTO Grant Veazey. The approach aims to improve reasoning across payer-specific rules, multi-step workflows, and regulatory nuances—limitations that general-purpose AI models often face.

Ensemble currently manages end-to-end RCM for more than 30 health systems nationwide. The partnership leverages Cohere’s enterprise AI capabilities, including secure on-premise deployments and full-stack control, ensuring confidentiality and compliance with healthcare data standards, according to Cohere Chief AI Officer Joelle Pineau.

The model is expected to support AI agents that enhance workflow efficiency, reduce claims denials, and increase revenue capture for health systems. "Every dollar saved or recovered in the revenue cycle can be reinvested into facilities and patient care," Veazey added.

The companies are also developing an RCM benchmark dataset to measure model performance, with plans to release the model in the second half of 2026. The initiative builds on a two-year data partnership between Ensemble and Cohere, focusing on real-world, implementation-first AI solutions for healthcare finance.

Health systems are increasingly exploring AI for RCM, with 80% piloting or implementing generative AI tools in 2025—a 38% increase from two years prior, according to HFMA and AKASA surveys. Ensemble’s LLM aims to meet this demand with a proprietary, domain-specific model tailored to the complexities of healthcare financial operations.

Stay tuned for more such updates on Digital Health News