How healthcare leaders can influence AI adoption

Artificial intelligence is reshaping healthcare—but separating hype from reality is critical. In a recent episode of CHIMECAST, the podcast of the College of Healthcare Information Management Executives (CHIME), Guidehouse Partner Erik Barnett shared practical insights for moving beyond theory to achieve measurable impact. In this interview with CHIME CEO Russ Branzell, he offered four key takeaways for healthcare leaders navigating AI adoption.

1. An enterprise strategy can help cut through noise about AI. 

Many healthcare organizations are racing to adopt AI—but without a clear strategy, they risk fragmented solutions and wasted resources. Barnett emphasizes starting small, building early wins, and scaling deliberately. Leaders must avoid the current trend of cobbling together dozens of point solutions without a clear understanding of how the pieces will work together. 

“The whole AI point solution versus enterprise strategy is one of the big conversations we continue to have,” Barnett told Branzell. A phased approach helps organizations manage costs while achieving measurable success—an important consideration for organizations that have limited funds but big aspirations.  

2. Governance is the foundation for success  

Successful AI adoption starts with governance. Barnett advises treating AI as both a distinct initiative and an integrated part of enterprise decision-making. Every use case should include a business case and performance tracking—especially as algorithms evolve daily. 

AI needs to be stress-tested, Barnett said. For example, a client that tested AI language translation against human translators determined that the AI solution failed to deliver accurate translation for certain languages. As a result, the health system opted to spend more time training the model while implementing limitations on its use.  

What separates successful adopters from those that flounder is governance and thoughtful strategy, he told Branzell. Because AI is still in its early stages, leaders should treat it as distinct from traditional technologies—while exploring how it can complement existing systems and their overall business strategy.  

“Use cases should be presented with business cases and tracking, because AI algorithms also evolve,” Barnett said. “There are new ones pretty much every day.” 

3. Trust comes from transparency, training, and realism. 

Leaders should not only evaluate the technology but also focus on how their teams will interact with it, facilitating adoption and support for people, not just processes. 

“We have to think about it differently,” Barnett said, noting that leaders need to align staff with organizational strategy while making it clear that AI is evolving quickly. “The days of the three-to-five-year strategy plan are gone.” 

Evaluating employee experiences with AI is important—it can help leaders understand to what degree AI is making employees’ lives easier, and help them understand obstacles to success, such as hallucinations or errors. That can help leaders get ahead of issues before scaling up.  

Ultimately, humans still need to be a part of the decision-making process. While thinking through how to automate administrative tasks, a nurse or care manager should still make final decisions. 

4. Capacity management is an overlooked use case for AI. 

Capacity management isn’t flashy, but it’s critical—and often overlooked as an area where AI can deliver real impact. Intelligent agents can streamline bed management and optimize patient placement and prioritization. 

AI agents can help make capacity management decisions by analyzing patient characteristics, comorbidities, licensing constraints, and physical layout data. Rather than placing every decision on nurse managers, AI systems can provide recommendations for clinicians to review and give them the opportunity approve, decline, or adjust, keeping humans in the loop while reducing cognitive load.  

Addressing capacity management challenges can unlock improvements across the board. “Patient length of stay is one of the key KPIs that is so critical to cost reduction, but so are patient experience and provider experience,” Barnett said.  

Execution is key to making enterprise strategy work 

The challenge isn’t generating ideas to use AI—it’s executing them. Success will require enterprise strategies, robust data infrastructure, and a commitment to incremental wins on the path to transformation. 

“My big bet is that our county hospitals, our community hospitals, and our academic medical centers will use AI to essentially provide an experience where patients are moved through the system faster, have the right information to make decisions, and that we are focused on helping cure diseases with AI.”