Epic VP of Data and Research Phil Lindemann
Photo: Epic
When Epic Systems opens its booth (#12713) on the show floor at the 2026 HIMSS Global Health Conference & Exhibition in Las Vegas next Tuesday, company representatives will have a lot to talk about as clients and potential customers stop by.
At the booth, almost assuredly decked out with fanciful sculptures and other pieces of art, conference attendees can sign up for details and demos about some of the many new products and priorities Epic has been focused on in the past few months.
There's its new AI Charting tool, of course, already at work in Epic's home state of Wisconsin, via Art, its new clinical assistant. There's also Emmie, its patient-facing chatbot; Penny, its automated rev cycle tool; and other agentic AI innovations.
But for a company that's been known over the years to invest as much as half of its operating expenses into R&D, you can be sure there's a whole lot more to showcase than just those tools.
We checked in recently with Phil Lindemann, Epic's VP of data and research, for a bit of a sneak peek into what the company plans to be highlighting HIMSS26.
Q. What new technologies will you be showcasing – introducing or previewing – at the show?
A. I'm particularly excited about Agent Factory, which will allow our customers to build and orchestrate AI agents that can reason, decide and execute steps autonomously. Our customers have so many great ideas for where AI can help. With a visual drag-and-drop builder, Agent Factory will help them to bring these ideas to life. We're just beginning to make Agent Factory available, and we're excited to see where it goes.
Another focus area is Art, our AI for clinicians. AI Charting by Art is already in use across multiple outpatient specialties, and the response has been great. But charting is just the beginning. Looking ahead, Art will make patient visits so much better.
As Art listens to the visit, it will pull up relevant information from the chart, surface insights from similar patients in Cosmos, and help the clinician and patient make decisions together – all in real time based on the discussion, without anyone having to pause and ask for help. We're also bringing the experience to new care settings. AI Charting for bedside nursing launches this month, with home care following in April.
And here's what I'm most excited about: In February, Curiosity launched for researchers in the Cosmos community. Curiosity is a new family of medical foundation models trained specifically on actual medical events in Cosmos. The models make predictions about what comes next in a patient journey.
Early research shows Curiosity performs as well or better than single-purpose models in predicting outcomes like length of stay, readmission or future disease risk. We're excited to work closely with researchers to continue evaluating Curiosity and bring its insights to the point of care in the future.
Q. Epic has introduced an array of new AI tools recently. What are the AI use cases you think hold most promise for your customers?
A. Today, most of our customers use Epic AI. They're seeing real, measurable gains.
Art has already saved thousands of hours in charting time. For example, in January alone, Art's Draft Hospital Course Notes saved clinicians over 1,300 hours on discharge summaries. Outpatient and Inpatient Insights, which summarize data from both internal and external sources, are used more than 16 million times per month.
Meanwhile, Penny – Epic AI for revenue cycle and operations – is transforming how billing and coding teams work. More than 240 organizations now use Penny to help with professional billing coding, with top performers reporting more than 20% fewer claim denials with Penny assisting with coding. Denial appeal letters are being created 23% faster with AI, too, across the community.
Our vision for Penny – and all our AI – is for it to become more autonomous. For example, Penny will complete straightforward coding sessions and drafting and submitting appeals independently within guardrails set by each organization.
Emmie, Epic's patient-facing AI, is already helping with billing questions, care plan follow-through and appointment scheduling through conversational SMS. For example, at Rush University Medical Center, Emmie has handled over 14,000 billing-related conversations with patients. With that has come a sustained 42% reduction in billing-related messages staff need to handle, all while patients get answers quickly and conveniently.
Emmie also provides patient-friendly, conversational follow-up reminders based on information in the patient's chart. These reminders help patients adhere to their care plans, and they have a 94% satisfaction rate!
Looking ahead, the biggest advancement is in agentic AI – AI that doesn't just assist but can reason, decide and take action within guardrails. Agent Factory is the platform behind this, and it's built on Epic's integrated system. This is important because just as humans need to move between domains like clinical, scheduling and analytics to complete their work, AI agents do too.
Q. What are some other strategic areas of focus for Epic this year?
A. It comes down to three priorities: connecting health systems around the patient, enabling AI across those connections and advancing the practice of medicine.
For the first of those, we continue to expand the healthcare network that connects patients, providers and payers. For example, we're advancing payer-provider collaboration through tools like Coverage Finder, Electronic Prior Authorization and ID Card Exchange – where 1.5 million insurance cards are digitally created every month.
Through Community Registries, we're helping health systems share and act on outcomes data across organizational boundaries. And with Aura, we're connecting specialty diagnostics, imaging, cardiac monitors and continuous glucose monitors so ordering is simpler, more results flow back into the chart, and clinicians get a fuller picture of each patient's story.
For the second, enabling AI across those connections, that's the story of Art, Penny, Emmie and Agent Factory, which we've covered. The key point is that AI becomes far more powerful when it can operate across a connected system.
And on the third priority, advancing the practice of medicine, that's where Cosmos and our clinical research tools come in. Cosmos now contains more than 300 million anonymized patient records and 17 billion healthcare encounters across four countries, with a fifth country currently installing it!
The Cosmos community has published more than 170 scientific articles, which have appeared in half of the top 100 peer-reviewed medical journals. One million clinicians are now on the Cosmos-enabled Look-Alikes network.
We're expanding point-of-care tools like Best Care Choices, which helps clinicians identify data-driven treatment options by comparing a patient's situation against outcomes from similar patients across the Cosmos dataset. Best Care Choices is available today for conditions like hyperlipidemia, hypertension and type 2 diabetes, with more on the way.
Clinical trials continue to be a focus for our health systems. Most of them offer research as a care option, and MyChart recruitment tools are now used more than 7 million times each year to notify patients of new opportunities. In collaboration with those health systems – and with life sciences companies – we're looking to get new treatments to patients even faster with our Discovery and Forward products.
We've started with Teleport Services, an opt-in program that's part of Discovery. It matches health systems with trial sponsors to connect patients with potential trials for new therapies. This could be a lifeline for patients that have exhausted all existing treatment options.
And now, we're working on Forward, our integrated clinical trials management system. It's integrated with the rest of Epic, including Discovery. It'll support end-to-end study management, from coordinating handoffs for study activation to tracking sponsor billing and participant progress – eliminating switching between systems, double-documentation and information silos. The first phases of development will be available starting in 2026, with continued rollout in 2027.
Q. What should visitors to the Epic booth be looking for and asking about at HIMSS26?
A. I know a lot of people are going to come for AI, and they should. But my advice is this: Don't just head straight for the AI section of the booth. Think about where you want to apply AI – documentation, patient engagement, revenue cycle, clinical trials – and go talk to those teams. The developers will be there. That's where the real conversations happen.
Two years ago, we told the industry what we were going to build. We built it. Now almost all our customers are using generative AI, and the outcomes are real and measurable.
I especially hope that organizations that aren't using much AI ask about how they can accelerate their own adoption. We can help! Launchpad is a program that helps organizations rapidly roll out generative AI features through guided, wave-based implementation. And our Operational Services team works directly with customers to solve operational challenges and build custom AI-powered automations tailored to their needs.
I'd also encourage people to explore Cosmos and Curiosity. We think that, together, they might fundamentally change how care decisions are made – moving from "what usually works" toward "what's most likely to work for you." That's what we're building toward, and the early results are encouraging.
And ask about Forward, our clinical trials management system, and EpicOps, our integrated ERP. These represent Epic's expansion into areas where health systems have been asking for better, more connected solutions. And we're gearing up to deliver.
Q. What are some questions or challenges you've been hearing recently from clients?
A. One of the most common challenges is simply knowing where to start. When you're understaffed – as healthcare is, across the board – picking which problem to tackle first is often the hardest part. We have hundreds of benchmarks across clinical and financial performance that help organizations see where their biggest opportunities are compared to their peers. When you're informed by that kind of data, it's much easier to focus your energy where it'll have the most impact.
Beyond that, organizational readiness is a consistent theme. Getting users comfortable with AI is as important as the tools themselves. The trustworthiness and integration of Epic's AI have helped drive adoption, and every organization is on its own journey.
Clinicians want reliability above all. If they're trusting AI, it needs to work essentially all the time. A recording failing, a workflow breaking or an incorrect order are deal breakers. So we're being really deliberate about reliability and usefulness – for example, as we released AI Charting by Art, our developers spent many hours side-by-side with the first clinicians to use it to make sure we got it right. We want to be sure Art makes clinicians feel like something has been taken off their mental plate so they can focus on patient care.
That's why the first release of AI Charting includes voice-driven preferences – so for example, if a clinician says that she prefers her HPI to be in bullet form, Art will apply that preference to the current note as well as future notes.
Something else we've learned is how much regional differences matter. For example, in the Netherlands, our teams learned that chart summaries needed special handling to never use the patient's name. That feedback came directly from clinicians there. The nuance is important, and our teams are on the ground listening and iterating with customers around the world.
Another big topic is what happens as AI moves from assistive to autonomous. The questions organizations are asking – how do we supervise AI agents, monitor them, QA them? – are exactly the right ones. There's still a lot of work we need to do in this area. Agent Factory gives us a repeatable platform to scale agentic capabilities responsibly as we and our customers learn more.
And of course, financial sustainability is on everyone's mind. With margins tight across the industry and rural hospitals at particular risk of closure, operational efficiency is directly tied to preserving patient access to care. That's driving interest in EpicOps – our integrated ERP – and in Penny's expanding capabilities on the revenue cycle side.
Q. Any other news of note you'd like to highlight?
A. We collaborated with Donate Life America on a feature that enables MyChart users to register as organ, eye and tissue donors directly within the app. That feature went live in May 2025, and already over 130,000 people have used it to register as organ donors. With over 100,000 people in the U.S. currently waiting for organ transplants, this is a meaningful step in helping people save lives.
On interoperability, we continue to expand what's available on Showroom and open.epic.com. In the past year, over 2,400 apps called a live API, and over 1,000 apps are listed on Showroom today.
And broadly: AI gives us the chance to rethink what caring for patients really looks like. Today, roughly half of patients see their doctor each year – and many of the rest probably need a nudge to get that cancer screening, make sure they're taking their blood pressure medication or follow up on a referral.
AI changes the equation. It expands a digital workforce that helps your doctor focus on the patients who need the most attention, while keeping everyone else healthy and engaged. That's what we're building together, and the organizations that lean in are compounding their advantage every day.
Epic will be in Booth 12713 at HIMSS26.
Mike Miliard is executive editor of Healthcare IT News
Email the writer: mmiliard@himss.org
Healthcare IT News is a HIMSS publication.
