What You Should Know
- The Milestone: At HIMSS26, Epic announces that more than 85% of Epic’s customer base is now actively utilizing its suite of artificial intelligence tools.
- Clinical ROI (Art): Epic’s clinician-facing AI, Art, is generating massive clinical and operational leverage. The ambient listening tool, Chart with Art, is now expanding into bedside nursing (live at Houston Methodist) and home care. Beyond charting, Art is actively driving clinical outcomes—such as boosting early lung cancer detection rates to 69% at The Christ Hospital by analyzing incidental radiology findings.
- Financial ROI (Penny): The revenue cycle AI, Penny, is autonomously attacking prior authorizations and claim denials. Summit Health cut prior authorization submission time by 42%, while other highly active systems have dropped coding-related denials by over 20%.
- Patient ROI (Emmie): The patient-facing AI, Emmie, is acting as a digital front door. At Rush University Medical Center, it drove a 58% reduction in billing-related customer service messages. Sutter Health is the first to go live with Ask Emmie, a conversational AI embedded directly inside MyChart.
- The Roadmap: Epic is moving aggressively into agentic AI and proprietary LLMs. The company announced Agent Factory, a platform for health systems to build and monitor custom AI agents, and Curiosity, a new family of medical foundation models trained on anonymized real-world patient records to predict patient journeys.
Epic Unveils ‘Agent Factory’ and Custom Foundation Models at HIMSS 2026
At the HIMSS 2026 conference, Epic revealed the staggering scale of that empire: more than 85% of its customers are now actively using Epic AI. The company has essentially deployed artificial intelligence across the majority of the U.S. healthcare system through a unified, three-pronged architecture: Art (for clinicians), Penny (for the back office), and Emmie (for patients).
Hard ROI Across the Enterprise
What separates Epic’s HIMSS presentation from standard tech-conference vaporware is the sheer volume of hard, operational ROI. The tools have moved past the pilot phase and are fundamentally altering hospital economics and clinical outcomes.
Agent Factory Launch
The company announced Agent Factory, a fully integrated platform that allows health systems to use a visual builder to create, customize, and deploy AI agents that can reason and act across workflows autonomously. Even more significantly, Epic is building Curiosity—a proprietary family of medical foundation models trained on anonymized, real-world patient records to predict disease progression, medication efficacy, and outcomes.
The Clinical Frontline: Epic’s ambient listening tool, Chart with Art, is aggressively expanding its footprint. After launching in outpatient specialties earlier this year, it became the first ambient scribe integrated into bedside nursing workflows (via a launch at Houston Methodist) and is rolling out for home care in April. But Art is moving beyond just taking notes. At The Christ Hospital, the AI is actively extracting incidental findings from radiology results to drive follow-up care, yielding a massive 69% early detection rate for lung cancer, compared to the national average of 46%.
The Revenue Cycle: On the financial side, Penny is dismantling the administrative friction that paralyzes hospital billing. Summit Health has cut submission times by 42%, with 92% of the AI-generated responses accepted without edits deploying AI against prior authorizations. Furthermore, health systems are seeing coding-related claim denials drop by more than 20%, effectively plugging a massive hole in their revenue leakage.
The Patient Experience: To deflect administrative bloat away from human staff, Epic has deployed Emmie. Acting as an intelligent, patient-facing triage layer, Emmie has delivered a sustained 58% reduction in billing-related customer service messages at Rush University Medical Center. At Sutter Health, patients are now using Ask Emmie directly inside MyChart to ask conversational health questions that are actually contextualized by their own medical records.
