Health System Shifts From Experimental AI to Operational Deployment
Aultman Health System, a three-hospital nonprofit based in Canton, Ohio, is moving beyond pilot projects to embed artificial intelligence across clinical and administrative operations. The shift reflects a broader recognition among hospital leaders that the current staffing and efficiency crisis demands a different approach.
Raza Fayyaz, chief information officer at Aultman, said the health system faces the same pressures as competitors: staffing shortages, complex workflows, and tight margins. "The status quo is no longer sustainable," he said.
Three operational pillars
Aultman's strategy rests on three components: pre-built AI tools, robotic process automation, and large language models that can handle reasoning-intensive tasks across the organization.
On the clinical side, the system uses ambient listening technology to reduce physician documentation burden, AI-first EHR workflows, and automated nursing systems. Internally, it is building custom models for patient discharge education and help desk support.
The supply chain represents the next frontier. Aultman is designing automated agents to match purchase approvals against invoices and track overall spending patterns. The goal is a closed-loop system that ties each purchase directly to measurable return on investment.
The real problem AI solves
Fayyaz emphasized that the core issue is not technology itself but administrative burden. "Our doctors and nurses didn't go to school to become data entry clerks," he said. "AI represents the first technology that promises to give the gift of time back to the bedside."
This framing matters. The health system views AI as a force multiplier for staff, not a replacement tool. The C-suite backs this position, treating AI as a strategic priority rather than an IT initiative.
Cultural requirements for success
Fayyaz's advice to healthcare leaders centers on treating AI as a cultural shift, not a technical project. He outlined four priorities:
- Purpose: Define why you are implementing AI. Without a clear vision tied to patient and staff outcomes, deployment fails.
- People: Address workforce anxiety directly. Leaders must publicly reject headcount reduction narratives to create psychological safety for adoption.
- Process: Identify where friction exists. Find champions-physicians, nurses, and administrative staff-who can spot problems AI can solve.
- Platform: Establish strong data governance and select tools that fit clinical needs while preventing unauthorized AI tools from proliferating.
Fayyaz also recommends that leaders start small. Building a simple agent to summarize daily calendars or emails using Microsoft Copilot or Google Gemini provides hands-on experience without high stakes.
"If we don't lead this, the AI will happen to us instead of happening for us," he said.
Aultman has established a data and AI governance framework to ensure all tools meet safety, ethical, and accuracy standards. This infrastructure is essential as the system scales deployment across multiple hospitals.
For operations leaders managing supply chain, workflows, or administrative processes, the operational AI strategy Aultman is executing offers a blueprint. The emphasis on AI agents and automation to eliminate manual work, combined with clear governance, reflects how mature organizations are approaching the transition from pilot to production. Operations professionals seeking to understand this shift may benefit from an AI learning path for operations managers that covers process optimization and workflow automation.
Your membership also unlocks:
