VERONA, Wisc.—As healthcare organizations rapidly adopt artificial intelligence technologies, electronic health record giant Epic aims to be the foundational layer, embedding AI and automation into clinical, operational and administrative workflows.
At its annual Users Group Meeting in Verona, Wisconsin this week, Epic unveiled a new suite of AI tools, new capabilities for clinical decision-making at the point of care and workflow automation, all building on initiatives announced at the 2025 UGM and also at industry conferences throughout the past year.
CEO Judy Faulkner, dressed as a cowgirl as part of the company's county fair-themed event, framed the three-hour executive address on Tuesday morning around the company's development of Ergo, what she described as "healthcare intelligence," a concept she teased at last year's UGM. Ergo combines the strengths that humans bring to the table (judgment, empathy and intuition) with AI's ability to find patterns across huge datasets in the EHR and to iterate quickly. "Plus, [AI] is always available, and it never gets tired," Faulkner said.
Ergo is built on a data foundation of the EHR, the patient's MyChart, Epic's healthcare network and Cosmos, the company's clinical research platform.
Epic is weaving generative AI and automation directly into its EHR, and the company is making a big push to highlight how its AI capabilities and copilot tools—Art, its clinical AI, Emmie, it's patient-facing AI and Penny, a copilot for revenue cycle—can all work together to function as an enterprise AI system for healthcare.
On the outpatient side, Faulkner said Ergo Visit is now live at Ochsner Health, as well as five physicians in Epic's Garden Plot collaborative. As part of the Ergo Visit, as Faulkner explained, prior to the patient visit, Art, Epic's AI assistant for clinicians, can provide a summary of the patient's record to get clinicians up to speed on the most clinically relevant points.
"Emmie talks with the patient in MyChart ahead of the visit, and then together Art and Emmy work to create a list of possible topics for the physician. Art creates summaries, key information for the healthcare team, and these summaries are in use now at over 300 healthcare organizations, and they get good feedback. Users tell us it summarizes information that helps them catch what might otherwise have been missed, especially things that come in through Care Everywhere that are coming from another healthcare organization, and they're amazingly quick," she told the audience.
Chart with Art, the company's AI-native charting tool, listens to the conversation between the patient and clinician, drafts the note and tees up other documentation for the clinician to review.
"Our goal was to redesign the office visit as we would build it if we were starting from scratch, with all of the technology we have available today," said Garrett Adams, senior vice president of R&D at Epic.
A big headline during Epic's presentation was the rollout of its Agent Factory, which is previewed at the HIMSS26 conference in the spring.
Agent Factory is Epic's platform for building, running and monitoring AI agents that can reason and act across workflows. "You can take it, shape it, make it," Faulkner said. "You can take one of our currently available 120 AI features out-of-tje-box. You can shape it for your organization, like equipping Emmie with a visitor guide, so it can help patients find the best places to park. That's shaping it, or you can make an agent yourself and share it with others."
Through Agent Factory, healthcare organizations can build something new without writing code. Early adopters are using Agent Factory today, according to the company, and it will be widely available in 2027. ECU Health built an AI agent to summarize transfer center requests, which improved handoffs and saved an estimated 20 hours a week, according to Epic.
As the healthcare AI market has become increasingly competitive with startups and health tech vendors pitching third-party tools, Epic is placing a big bet on its ability to embed tools deeply into EHR workflows.
Speaking on stage, Seth Howard, vice president of R&D, stressed that the new tools and capabilities highlighted on Tuesday were not separate features. "Under the hood, it's a single integrated system: Art, Emmy, Penny, Agent Factory, Curiosity, the healthcare network, together with you as one team working to make healthcare better for everyone."
Epic also is leveraging is clinical research database, Cosmos, to build a generative AI model, called Curiosity, that uses real-world patient data to forecast outcomes such as hospital readmission and stroke risk. Cosmos combines clinical data from a community of Epic Systems health systems representing over 310 million patients and 21 billion medical encounters, according to Epic.
Twenty healthcare organizations are working to validate Curiosity, Faulkner noted, and the predictive tool will be available to any organization contributing to Cosmos.
"Curiosity learns from the patient charts in Cosmos, so Curiosity can present possible answers for you and for your patient, even when there's no published study to review. We think Curiosity predictions will become extremely important for healthcare organizations as you care for your patients," Faulkner said.
As a generative AI model trained on the Cosmos dataset, Curiosity can answer clinical questions at a population level as well as about a specific patient's health trajectory, executives said.
Patients are quickly turning to AI tools to supplement their healthcare as every country faces healthcare constraints and access issues, Carl Dvorak, a top Epic executive, told the audience on Tuesday. Epic wants to work with health systems and providers to leverage AI safely and responsibly while also keeping up with the pace of innovation, he noted.
"When I think of AI, the thing that comes to mind first is we've got to find a way to move faster, especially given that patients are as broadly adopting this right now. We've got to find a pathway to our AI utopia before patients become frustrated with the health system and frustrated that you can no longer be trusted as their primary source of medical information," Dvorak said.
"We're going to have to figure out who to trust. We have to trust them, and we're going to have to move. So that means for you guys, a call to action here: get engaged and stay involved with regulatory bodies. And equally as important, be leaders internally. Let's be leaders and let's find our path through this, and not find another reason to wait, because if we don't get through it, others will. We've got to find a path through that so that we can keep pace with the future of AI," Dvorak said.