Crucial Conversation: AI and the EHR | What Comes Next for Healthcare Delivery

Artificial intelligence continues to dominate conversations across healthcare, but turning its promise into meaningful impact requires more than adopting the latest technology. It requires solving the right problems, earning clinician trust, and implementing solutions that enhance—not complicate—the delivery of care.

Moderated by Andrew Kerr, CEO of FortyAU, this Crucial Conversations session brought together Dr. Todd Clark, Medical Director at athenahealth and practicing emergency physician, and Dr. Avinash Bachwani, Corporate Vice President and Chief Medical Information Officer at Community Health Systems, to discuss where AI is already creating value, the challenges organizations continue to face, and what responsible adoption should look like moving forward.

Moving Beyond the "AI Show Pony"

Kerr opened the discussion by reflecting on the evolution of the electronic health record. Originally built to support billing and administrative functions, EHRs have steadily become central to care coordination, clinical decision-making, and workflow management. Today, AI represents the next stage of that evolution—but only if it delivers measurable value.

"I hope we've gotten past what I call the AI show pony," Kerr said, explaining that healthcare organizations have largely moved beyond experimenting with AI simply because it is new. Instead, the focus has shifted toward identifying practical use cases that improve workflows, reduce administrative burden, and produce meaningful business outcomes.

Drawing from projects his team has implemented over the past year, Kerr highlighted examples ranging from AI-powered gap-in-care tools and Chrome extensions that improve ambulatory workflows to enterprise-wide nurse handoff summaries that standardize patient information during shift changes. In each case, the objective remained the same: embed AI into existing workflows in ways that simplify work for clinicians rather than disrupt it.

Ambient Documentation Sets the Standard

When the conversation turned to successful AI implementations, all three speakers pointed to ambient documentation as the clearest example of technology delivering immediate value.

Clark described ambient documentation as transformative for clinicians because it removes one of the profession's greatest administrative burdens without fundamentally changing how providers practice medicine.

Rather than balancing patient conversations with extensive documentation, clinicians can focus their attention on the patient while AI captures the encounter and generates the clinical note.

For Clark, the impact has been personal. He shared that the documentation burden in emergency medicine had become significant enough that he had considered stepping away from clinical practice. Ambient documentation changed that outlook by allowing him to spend more time practicing medicine and less time completing paperwork.

Bachwani echoed that experience from the health system perspective. Community Health Systems has already surpassed half a million AI-supported encounters across inpatient and outpatient settings, demonstrating both the scalability and value of the technology.

Even so, adoption remains between 35% and 40% among clinicians.

That statistic underscored one of the panel's recurring themes: even when technology works, successful implementation depends on clinician confidence, thoughtful change management, and organizational support.

AI Should Reduce Cognitive Burden—Not Clinical Judgment

improving clinical workflows and replacing clinical decision-making.

Bachwani explained that AI excels at computational tasks such as recognizing patterns in medical imaging, processing vast amounts of information, and summarizing complex patient records. Clinical reasoning, however, remains fundamentally different.

"Mathematical reasoning" and "clinical reasoning," he noted, are not the same. Diagnosis requires context, experience, communication, and judgment that extend well beyond pattern recognition.

Clark agreed, emphasizing that physicians do not want technology making decisions on their behalf. Instead, AI should eliminate the administrative tasks surrounding patient care while allowing clinicians to focus on the work that only they can do.

As he explained, clinicians still want to solve difficult cases, exercise their judgment, and build relationships with patients. AI is most valuable when it removes the obstacles that keep them from doing exactly that.

Responsible Adoption Requires More Than New Technology

Beyond individual use cases, the discussion focused heavily on how organizations should approach AI adoption at scale.

Bachwani explained that Community Health Systems has evolved from implementing individual AI solutions to building a broader enterprise AI ecosystem supported by governance, steering committees, and clearly defined evaluation processes.

Rather than introducing disconnected tools across the organization, the goal is to create consistent experiences for clinicians while ensuring every solution aligns with organizational priorities and measurable outcomes.

Clark described a similar philosophy. The organization has adopted an "AI-first" mindset by training employees across the company while embedding AI capabilities directly into the platform instead of relying solely on third-party applications. At the same time, he emphasized that clinicians should retain control over how and when those tools are used, recognizing that comfort levels with AI continue to vary widely.

Throughout the discussion, Kerr returned to a broader point: successful innovation requires balancing speed with trust. Healthcare organizations cannot adopt technology at the pace of other industries when patient safety, privacy, and clinical outcomes are at stake.

Preserving Clinical Expertise for the Next Generation

The panel also explored one of AI's less-discussed risks: overreliance.

Bachwani pointed to AI-generated differential diagnoses for medical residents as one example. While the technology can produce valuable suggestions, relying on those recommendations too early in medical training could prevent future physicians from developing the clinical reasoning skills essential to independent practice.

Clark expanded on that concern, noting that clinical decision support tools should be introduced thoughtfully for trainees. Experienced physicians may benefit from AI surfacing additional information, but allowing AI to perform too much of the reasoning risks de-skilling future clinicians.

The conversation extended beyond medical education. Bachwani referenced AI-assisted endoscopy, where physicians who routinely relied on AI-supported cancer detection demonstrated lower detection rates once those tools were removed—illustrating that even experienced clinicians can become overly dependent on technology if its role is not carefully managed.

Looking Ahead

As the discussion concluded, the panel agreed that while AI often receives outsized attention in public discourse, healthcare is still in the early stages of realizing its long-term potential.

Rather than expecting AI to replace physicians or transform care overnight, the speakers envisioned a future shaped by hundreds of practical improvements that quietly eliminate friction throughout the healthcare system.

For Clark, that future lies in countless small efficiencies—surfacing the right information at the right time, simplifying documentation, and organizing increasingly complex patient records.

For Bachwani, the greatest opportunities extend even further into areas such as disease detection, medical imaging, and computational research, where AI's ability to process enormous volumes of information could accelerate breakthroughs that were previously out of reach.

Kerr closed the conversation by reminding attendees that successful innovation begins not with technology, but with people. The organizations that realize the greatest value from AI will be those that prioritize clinician workflows, earn trust through thoughtful implementation, and deploy technology with a clear understanding of the problems they are trying to solve.

About the Nashville Health Care Council

The Nashville Health Care Council strengthens and elevates Nashville as The Healthcare City. With a $68 billion economic impact and 333,000 jobs locally, Nashville’s healthcare ecosystem is a world-class healthcare hub. Founded in 1995, the Council serves as the common ground for the city’s vibrant healthcare cluster. The Council offers engagement opportunities where the industry’s most influential executives come together to exchange ideas, share solutions, build businesses and grow leaders.