AI exposes chaos when fragmented data and broken workflows become more visible during automation, says Michael Vipond of ServiceNow.
LAS VEGAS – Artificial intelligence is often discussed in the context of diagnostics and clinical decision-making, but some of the most immediate gains in healthcare may come from improving operations behind the scenes.
AI is increasingly being deployed to tackle the administrative burden that consumes a significant portion of clinicians’ and staff members’ time.
At the HIMSS Global Health Conference & Exhibition session “Reinventing Healthcare Operations: AI That Helps Your Team Reclaim Time for Patient Care,” Michael Vipond, head of Healthcare Provider Go-to-Market at ServiceNow, said many health systems are discovering that AI initiatives quickly reveal deeper operational challenges.
“AI exposes chaos,” Vipond said. “Fragmented data environments and broken workflows often become more visible once organizations begin experimenting with automation. The intelligence is the easy part with foundational LLMs.”
However, without addressing underlying systems and processes, AI tools struggle to deliver meaningful value. That reality is one reason some early AI pilots have failed to gain traction.
According to Vipond, projects that never integrate into day-to-day workflows risk becoming little more than “very expensive advice.”
Successful deployments require organizations to connect AI insights directly to operational actions through structured governance and clear oversight.
For health systems moving from experimentation to operational adoption, building trust and driving workforce engagement have become just as important as the technology itself.
Teresa Incesi, director of IT at Northwestern Medicine, said cultural resistance was one of the biggest barriers when the organization began expanding its AI initiatives.
“Getting adoption was our No. 1 challenge,” she said.
Leaders focused heavily on education and communication to help staff understand how AI tools could support their work rather than replace it.
The organization started by examining how information was structured across internal systems and identifying opportunities to improve search and knowledge management capabilities.
From there, teams evaluated potential AI use cases across clinical, administrative, HR and finance operations.
“We knew we had to get ahead of AI,” Incesi said.
A key focus was identifying high-volume workflows where automation could produce immediate benefits. Clinical departments were particularly active in proposing new applications, including tools to help surface radiology findings and assist with incident management.
The IT organization also applied AI to internal service management processes, including automated resolution note generation and support workflows.
Many of these initiatives were built around core enterprise platforms such as Microsoft technologies, Epic and ServiceNow, with additional custom AI models developed to support specific operational needs.
Incesi said education played a central role in encouraging adoption. IT leaders worked to demonstrate how AI tools could reduce administrative workload and allow clinicians to focus more time on patient care.
“The key is helping them understand AI will assist them and not replace them,” Incesi said.
Trust was reinforced by involving staff members early in pilot programs and maintaining transparency about what the technology could and could not do. Teams were invited to test new tools in real workflows and provide feedback before broader deployment.
“That’s how you develop evangelists,” Incesi said.
As those early users share positive experiences with colleagues, organizations can gradually expand adoption across departments.
For Northwestern Medicine, adopting a platform-based approach has also been essential for scaling AI capabilities. Standardizing on widely used enterprise systems allows teams to reuse skills and integrate new AI services more efficiently.
“The platform approach was really the only decision I had,” Incesi said, noting that aligning AI development with existing platforms helps organizations scale innovation without dramatically increasing technical complexity.