Transforming healthcare workforce models for the AI era

Redesigning the workforce

A common misconception is that AI’s primary value comes from replacing clinical expertise. The experience of healthcare organizations is pointing in a different direction.

Across nursing, diagnostic imaging, medication management, outpatient services and other clinical functions, healthcare organizations are finding opportunities to separate administrative processing from clinical judgment.

Joe Ranzau, Partner of Transformation at Grant Thornton, said AI-supported tools are put to good use when they can perform initial analyses while clinicians review results and make final decisions. During a recent eye exam, he saw technology automatically assessing imaging scans and generating preliminary findings before the physician reviewed them. The result was a more efficient workflow that still preserved clinical oversight. It also allowed portions of the process to be performed by a technician while the physician focused on reviewing results and applying clinical judgment.

“AI can enable highly trained clinicians to focus on reviewing results and applying their expertise,” Ranzau said. “Ultimately, AI should be used in healthcare to optimize existing processes.” In practice, AI can analyze data, flag anomalies, measure KPIs related to cost, quality and turnaround times, and generate recommendations, while clinicians provide the validation and judgment needed to act on those insights.

As Scanlan notes, physicians are often most receptive to AI when it is positioned as an enhancement rather than a substitute. “When AI reduces administrative burdens and gives clinicians more time with patients without diminishing professional judgment, that's when people truly embrace the technology," she said. 

Use cases delivering results

Some of the most promising workforce applications involve eliminating repetitive tasks that consume valuable clinical time.

Anichkin cited exception reporting in healthcare as an example of AI’s ability to improve operations. In many healthcare settings, exception reporting happens through nurses manually reviewing extensive reports to identify missed service-level agreements, medication delays, supply shortages, equipment failures or other issues that require follow-up. AI can automatically analyze those datasets and surface only the cases that require attention.

“AI can handle this analysis and surface the exceptions, allowing staff to focus on decisions and actions instead of spending time reviewing reports and performing manual calculations,” Anichkin said.

Other opportunities include automating portions of EHR documentation, streamlining radiology workflows, supporting diagnostic imaging analysis and reducing duplicate administrative work across clinical systems. In diagnostic imaging, for example, automation can reduce the manual steps required to move information between systems and prepare studies for review, helping staff focus more time on patient-facing activities.

The opportunity extends across the healthcare enterprise. From nursing operations and medication administration to radiology, outpatient services and care coordination, organizations are identifying opportunities to redesign workflows, improve visibility into operations and reduce process inefficiencies. 

Those questions can help shift AI planning from isolated technology pilots to practical operating-model redesign.

Choosing which processes should be automated

Successful healthcare transformation starts by identifying where AI should and should not be applied. Operational assessments have shown that some workflows may appear ripe for automation on paper but serve important human and cultural functions in practice. Nurse shift handoffs are one example. Although portions of the process can be automated, those interactions help fellow RNs convey situational awareness, transfer knowledge and preserve continuity of care.

Other activities involve more than information transfer because they reinforce accountability, trust and professional judgment. Those processes require a thoughtful balance between automation and human involvement.

Scanlan said that whenever considering applying AI use in practice, ensuring patient safety must remain the guiding principle. Organizations can identify many places where AI could assist in monitoring, analysis, or operational decision-making, but not every opportunity should be pursued first. The most successful implementations typically begin with lower-risk administrative processes where organizations can generate value, reduce clinician burden, and demonstrate results without introducing additional clinical risk. As frontline teams see the technology improving their day-to-day work, trust grows, and organizations can gradually expand into more complex and clinically sensitive use cases.

Managing change for AI workflow success

Healthcare leaders should focus on selecting the right AI platforms, but implementation success depends far more on designing processes around them. Many healthcare workflows were developed and instituted long before AI existed. Simply inserting a new tool into an existing process doesn’t automatically improve outcomes and sometimes can create additional complexity.

Effective AI adoption begins with detailed process mapping and stakeholder engagement. Healthcare leaders must understand how work is performed today, identify where bottlenecks occur and determine how responsibilities should change in the future state. Clinicians, operational leaders, technology teams and quality leaders all need a voice in that conversation.

Ranzau emphasizes that change management is as important as the technology itself when it comes to AI implementation.

“Successful AI adoption starts with helping clinicians understand why AI is being introduced, how it will work, and what their role will be in the new process,” Ranzau said.

Anichkin said, “AI implementation success comes from improving the end-to-end process and making work easier for the clinicians and staff who rely on it.” If a new AI-enabled process creates confusion for downstream users, increases complexity or requires more effort to interpret outputs, Anichkin said, the organization has not really improved performance.

Human-centered healthcare

Healthcare organizations considering AI investments should focus first on helping clinicians do what only they can do. Leaders should ask which administrative burdens can be safely removed, where physicians and nurses are losing valuable time, and how technology can support better patient care.

For healthcare executives, the business case extends beyond labor savings. AI-enabled workforce redesign can expand clinical capacity, improve retention, strengthen quality, and help organizations do more with limited resources.

The organizations seeing the greatest return from AI are treating it as a workforce and operational-model strategy rather than a standalone technology implementation. By redesigning processes around AI, healthcare organizations can expand capacity, improve workforce sustainability and enhance patient access while preserving the clinical judgment that remains central to care delivery.