LAS VEGAS—Epic is ramping up more artificial intelligence capabilities and features as it also touts how its AI tools drive measurable outcomes beyond just faster documentation time.
Health systems are reporting earlier diagnoses, fewer denials and improved patient experiences, the company said.
At the 2026 Healthcare Information and Management Systems Society (HIMSS) Global Health Conference & Exhibition, which kicked off Monday, Epic teased its future AI road map with new features across clinical, patient-facing and operational workflows.
The electronic health record giant plans to release a fully integrated platform for creating and monitoring AI agents that reason and act across workflows, the company announced Tuesday. With Agent Factory, organizations will be able to customize agents with a visual builder, equip them with local policies and knowledge bases and deploy them on their own timeline, Epic said in a press release.
At the J.P. Morgan Healthcare Conference in January, Phil Lindemann, Epic's vice president of data and research, previewed Agent Factory as a platform to enable organizations to build their own AI agents within Epic.
"Epic is going to build an entire library of on-platform agents that are just ready to go, that are baked into the system. They're completely integrated into the user's experience. But we have some of the most innovative health systems in the world who said, 'Epic, you need to give us a playground, you need to give us a sandbox, where we can invent, we can reimagine.' So that's what Factory is intended to be," Lindemann told Fierce Healthcare at JPM.
Epic has a number of new AI features in the works. The company is set to roll out conversational AI for clinicians as part of its Art clinical assistant, which includes an AI scribe tool. During the patient visit, the conversational AI will surface answers to questions that the clinician and patient ask—the AI tool can handle direct questions like "Art, how is this patient’s blood pressure trending?" It can also answer inferred questions based on conversational context, Epic said.
During its UGM in August, the company unveiled its AI scribe solution, called Art for Clinicians, along with a generative AI copilot for revenue cycle, called Penny, to help with coding and denial appeals. It also unveiled an AI solution for patients, called Emmie, to help patients schedule medical visits and then prepare for them.
Building on this, as part of a new AI feature, Art and Emmie will build collaborative visit agendas together to bring questions and priorities from the patient and clinician into one view, according to Epic.
Another new feature is a diagnosis check built within Art that will help clinicians with differential diagnosis by analyzing patterns in patient symptoms, history and outcomes and suggesting potential next steps based on real-world patient data.
Epic also is building on Curiosity, its family of new gen AI models, with plans to create a new family of medical foundation models trained on anonymized real-world patient records. The models help predict the next medical event across diagnoses, medications, procedures and outcomes.
More than 85% of Epic’s customers now use Epic AI, the company said. As health systems and providers place their bets on specific use cases for AI, conversations have shifted from theoretical potential to measurable outcomes and return on investment.
"AI is starting to show real outcomes. There's obviously some hype, but real things are starting to happen," Lindemann said at JPM in January.
Epic is bringing into focus key outcomes from its customers as AI cuts down paperwork, improves diagnosis and reduces costs.
Clinicians at multiple organizations are completing discharge summaries 20% to 30% faster with Art’s draft hospital course notes, the company said. At Riverside Health in Virginia, clinicians using Art’s inpatient insights are spending up to 32% less time on documentation and communication tasks.
The use of AI also is helping patients get treatment faster, Epic contends. At The Christ Hospital, Art extracts incidental findings from radiology results and drives follow-up, delivering an early detection rate of 69% for lung cancer versus the national average of 46%.
In March, Houston Methodist became the first organization to use Chart with Art for bedside nursing workflows. Epic plans to release Chart with Art for home care workflows in April.
Organizations also reported notable improvements in prior authorization by deploying Epic’s AI for revenue cycle and operations. At Summit Health, the use of Penny, Epic's revenue cycle AI tool, helped cut medication prior authorization submission time by 42%, and 92% of AI-generated responses are accepted without edits.
At healthcare systems most actively using Penny, coding-related denials have dropped by more than 20%—preventing revenue loss due to claim rejections and reducing the need for staff to spend time reworking claims, Epic executives said.
Epic's AI for patients, called Emmie, quickly gives patients answers to straightforward questions and reduces health system workload. At Rush University Medical Center, Emmie delivered a sustained 58% reduction in billing-related customer service messages.
Emmie also can tackle appointment scheduling. At Ochsner Health, patients have rescheduled more than 14,900 appointments so far with Emmie, making scheduling more convenient and saving almost 750 hours of staff time.
Sutter Health collaborated with Epic to become the first organization live with Ask Emmie, a conversational AI assistant in MyChart that answers patients’ health questions within the context of their medical record. Sutter and Epic worked closely on testing and refining it ahead of launch. That conversational AI assistant is currently available to an initial group of users, with broader availability planned in the coming weeks, Epic said.