Ensemble, a leading end-to-end revenue cycle managed services partner for healthcare organizations, has announced an expanded collaboration with Cohere, a security-first enterprise AI company, to develop the healthcare industry’s first revenue cycle management (RCM)-native large language model (LLM). The initiative represents a significant step forward in applying purpose-built AI to complex healthcare financial operations.
Unlike many current solutions that rely on adapting general-purpose LLMs through prompt engineering, this collaboration focuses on building a fully customized AI model. The model is being designed from the ground up using deep RCM insights, operational expertise, and structured workflows to deliver consistent, measurable performance for healthcare systems.
Traditional AI approaches often attempt to introduce RCM logic during inference using extensive contextual inputs. However, this method can increase operational costs, strain model reasoning, and limit accuracy—particularly when dealing with payer-specific rules, regulatory nuances, and multi-step workflows that define healthcare revenue cycle processes.
To address these limitations, Ensemble and Cohere are adopting a real-world, implementation-first strategy. The new RCM-native LLM is being developed using Ensemble’s extensive operational data and expertise, fine-tuned on real-world RCM tasks, and embedded into AI agents that support end-to-end workflows—from patient intake through account resolution.
Also Read: Extend Expands ERP Integrations to Modernize Bank-Led Expense Management
“For more than a decade, Ensemble’s domain expertise has powered our clients’ financial performance and award-winning RCM results. What we’re building with Cohere elevates this advantage,” said Ensemble president + CEO Judson Ivy. “Our associates’ operational knowledge, well-defined processes and insight into payer behavior are shaping how models learn and solve problems that will help us reduce friction across the patient financial journey and continue delivering the trusted results health systems count on.”
This integrated approach establishes a purpose-built intelligence layer specifically tailored for healthcare RCM. The model is designed to understand complex clinical, financial, and regulatory language while navigating intricate payer requirements and documentation processes. By closely mirroring the workflows and decision-making patterns of experienced RCM professionals, the system is expected to significantly enhance operational efficiency and accuracy.
One integral part of the project is the enhanced focus on data privacy and compliance. Neither the model is trained on individual identifiable client data nor on protected health information (PHI). It utilizes industry-wide aggregated insights, documented processes, patterns of payer behavior, as well as synthetic datasets generated from certified, de-identified sources in a HIPAA-compliant manner. Significantly, the RCM-native model does not aim to replace electronic health record (EHR) systems.
It actually upgrades EHR-based workflows by incorporating context intelligence and task-specific guidance. It facilitates decision making, documenting, and following up actions, enabling users to meet requirements that are beyond traditional EHR capabilitiesstill, without altering existing system configurations. With this project, Ensemble and Cohere are pioneering AI-enabled health care operations by providing a solution that is secure and scalable, resulting in improved accuracy, minimized administrative workload, and increased support for financial and clinical outcomes in health systems.
