If you’ve spent any time at an industry conference, the hotel bar next door, or with your Epic team, you’ve noticed something simultaneously fascinating and frightening: artificial intelligence is no longer a future-state conversation or a separate implementation. AI is here, and it’s not a bolt-on. (Does anyone else quietly play AI buzzword bingo at work? You won’t even need the free space.)
Ambient documentation. AI-assisted workflows. Predictive capabilities. Generative tools embedded directly into clinical and operational processes. The industry’s largest EHR vendor has made it clear that AI is not a side project. It’s a strategic direction that will be weaved into the software and eventually a native “cruise control” that is simply a transparent part of the EHR.
And yet, adoption of Epic AI features is inconsistent across the Epic community. Some organizations are racing ahead, deploying new Epic AI capabilities almost as quickly as they are released. Others are moving much more cautiously, adopting only a handful of features or avoiding them altogether. The question is: why?
It isn’t because Epic’s AI capabilities are unavailable. It isn’t because the technology doesn’t work. While financial concerns are certainly part of the conversation, I’d argue there is ample ROI evidence suggesting that, if adopted well and at scale, most AI features can pay for themselves quickly.
I argue the answer is much simpler. As usual, the difference is people.
We’ve Solved the Technology Problem
For decades, healthcare IT organizations staffed Epic teams around a relatively consistent model. The strongest analysts were often those with deep application expertise. They understood the Chronicles data structure and showed off Lookitt to their cube mates. They knew their application inside and out. They displayed their certifications on the wall and had years of experience configuring, maintaining, and optimizing increasingly complex environments.
To be clear, there is nothing wrong with this model. In fact, it was exactly what the industry needed—in the 2010s. Healthcare organizations spent years implementing EHRs, replacing legacy systems, integrating acquisitions, supporting regulatory initiatives, and optimizing increasingly sophisticated technology platforms. Deep application expertise was critical. Organizations needed analysts who could build, configure, troubleshoot, and support. The industry got very good at developing these professionals.
The problem? AI adoption is a fundamentally different challenge.
Certifications Won’t Drive Adoption
Several years ago, I wrote that healthcare organizations should reconsider some of the hiring profiles they use to attract and evaluate talent. One of the arguments I made was that EHR certifications—while valuable—don’t necessarily tell the entire story. They are often a ticket to the interview, not a guarantee of success. The skills that determine project outcomes frequently extend beyond technical knowledge and into communication, change management, operational alignment, and strategic thinking. I believe that observation becomes even more relevant in the age of AI.
Successful AI programs are rarely limited by whether someone can turn on a feature. More often, they are constrained by questions such as:
- Which AI capabilities should we deploy first?
- How do I calm my nervous CMIO?
- How do we prioritize competing opportunities?
- How do we measure adoption?
- How do we communicate value to providers and executives?
- How do we create an AI-first culture without creating AI fatigue?
These are not build questions. They’re leadership questions.
Consider ambient documentation as an example. Most organizations now have providers who are actively using these capabilities and realizing significant benefits. They also have providers who refuse to use them. Some cite concerns about workflow changes. Others claim the generated documentation doesn’t “sound like them”. (I didn’t know my visit note from the men’s clinic was an exercise in creative writing, but I digress.) Still others simply prefer established habits.
Whether those concerns are valid is almost beside the point. The technology itself is no longer the primary obstacle. Adoption is.
Introducing the Epic AI QuarterbackTM
At Divurgent, we’ve observed a pattern among organizations that are successfully accelerating AI adoption.
Many have, intentionally or unintentionally, created a new type of role—one that doesn’t fit neatly into traditional healthcare IT job families.
We’ve given that role a name: The Epic AI QuarterbackTM.
The Epic AI Quarterback is not a traditional analyst, nor is it simply a project manager, program manager, strategist, or change management leader. It’s a hybrid role that combines characteristics from all of them.
The Epic AI Quarterback possesses deep Epic knowledge and understands workflows across clinical, operational, and revenue cycle domains. They probably hold or held multiple certifications but don’t display them proudly anymore. They’ve figured out that deploying a modern EHR in the age of AI has very little to do with passing open-book tests.
Instead, they now understand the organization’s strategic priorities. They can communicate effectively with executives. They know how to develop business cases and articulate ROI. They understand organizational change. Most importantly, they relentlessly drive adoption. This role serves as the central leader coordinating AI initiatives, aligning stakeholders, tracking outcomes, and removing barriers to success.
Think of this individual less as a builder and more as an orchestrator.
They are constantly asking:
- What AI opportunities are available today?
- Which capabilities align with organizational priorities?
- What barriers are preventing adoption?
- What is the measurable return on investment?
- What is the cost of doing nothing?
- How do we sustain momentum after go-live?
Those questions ultimately determine success far more than whether you set ORD 30 right.
If your organization is serious about accelerating Epic AI adoption, it’s time to think beyond traditional staffing models and consider an Epic AI Quarterback.
Why Traditional Roles Aren’t Enough
This is not an indictment of existing Epic analysts. In fact, expecting a traditional application analyst to independently fulfill this function is unfair. Most analysts were trained and developed in a different era with different expectations. Many were hired because they excelled at application configuration, testing, support, optimization, and technical problem solving. Those capabilities remain incredibly important.
But AI transformation requires additional skills that many healthcare IT organizations have never formally developed or recruited for. Consider:
- Executive communication.
- Business case development.
- Cross-functional strategy.
- Value realization.
- Organizational change management.
- Enterprise prioritization.
These skills don’t always emerge through traditional analyst career paths, yet they are now essential. I argue, with excitement, that the talent pool for the Epic AI Quarterback role is larger than it may seem. Our industry is full of former analysts-turned-PMs with leadership drive. There are armies of people who have “done it all” and are coasting at $100+ an hour building SmartTexts or working tickets because nothing has excited them since the Obama administration.
Indeed, a common faux pas I see is to hire immediately to the top, such as a Chief AI officer. I’m open-minded but admittedly suspicious of this role. I’m yet to meet a healthcare Chief AI officer who can explain to me what steps he or she would take in Epic to solve problems like In Basket overload, note bloat, pajama time, and others. They’ll talk to me about data science, LLMs, and cover the basics, but often call their analysts when we reach the “what to do” stage. Hence the problem I’m seeing.
To healthcare CEOs: please, save your money until you’ve nailed the basics.
The Most Important AI Investment May Not Be Technology
Healthcare leaders often ask which AI capability they should deploy next. It may be a useful question, but perhaps the more important question is: Who is responsible for ensuring adoption after deployment?
Because technology alone does not create transformation. People do.
The organizations realizing the greatest value from Epic AI investments are not necessarily those with the most advanced technology, most maturity, or the most money. They are often the organizations with the right person(s) driving strategy, alignment, adoption, and execution. That person increasingly looks like an Epic AI Quarterback.
Your Best Placement Yet
The healthcare industry is still defining what this role looks like. The title isn’t standardized, and the job descriptions are still evolving. Many organizations don’t even realize they’re looking for it or need it. Still, the role is real. We’ve seen what works. We’ve seen what stalls. We’ve seen the difference between organizations that embrace AI as a strategic capability and those that struggle to move beyond pilot projects. If your organization is serious about accelerating Epic AI adoption, it’s time to think beyond traditional staffing models. The future belongs to professionals who can bridge technology, strategy, operations, and change. At Divurgent, we call that person the Epic AI Quarterback. It may be the most important role you never knew existed.
