AI Governance Needs ROI. Here's How Leaders Are Prioritizing

At Baptist Health, AI governance demands a deep collaboration across IT and clinical teams, strict contract metrics, and essential exit clauses to mitigate risk and ensure a verifiable ROI.

For Steven Kos, Senior Director of Revenue Cycle Application Support at Baptist Health, the era of operating in silos is over.

Kos was a participant in this year’s HealthLeaders Revenue Cycle Technology Mastermind program, which brought together RCM executives from roughly a dozen health systems for a series of virtual and in-person round-tables to discuss key benefits and challenges.

Kos, who brought a critical IT perspective to a discussion that has traditionally focused more on financials, says his message is clear: AI acceleration is forcing a necessary collision between clinical operations, ambulatory services, IT, and the revenue cycle.

"That is absolutely required for us to transition forward because it is an integrated solution," Kos notes. "AI only complements that – it expedites the necessity for us to work together because it accelerates change".

How to Govern AI Technology Contracts

The rise of AI has been accompanied by the need for governance structures over automation. At Baptist Health, this meant establishing an "AI Institute" to manage a growing portfolio new technologies.

For automation projects to move forward, they must have clearly defined goals and receive approval from the AI Institute.

This governance model has a direct impact on contracting, according to Kos. Solutions are not approved unless there are strong contracts supporting Service Level Agreements (SLA). Crucially, Baptist Health insists on contract clauses that allow them to exit if the projected ROI is not achieved.

Pilot programs must demonstrate potential for ROI, Kos says. And then “we have to take it back to AI governance and show that metrics have been met before we allow the pilot to become a final contract.”

Targeting High-Impact Areas.

Baptist Health is currently focusing its technology roadmap on specific, high-value targets for the coming year:

• Pre-Bill Optimization: The organization is deploying an AI-based platform to improve DRG accuracy and case mix prior to billing, with the goal of reducing denial rates associated with unsupported documentation.
• Autonomous Coding: Baptist is investing in autonomous coding for professional billing to reduce reliance on third-party labor.
• Denials Management: The team is implementing AI to automate appeal letters to payers.

Vendor Reality Checks

Despite the excitement around AI, Kos maintains a grounded perspective toward the vendor landscape. While the number of solutions is expanding, he says, there is currently an unmanageable surplus of choices because market consolidation has not yet occurred.

Furthermore, revenue cycle leaders should be mindful of the "change management gap." From an IT perspective, vendors often underestimate the operational friction involved in deployment.

"They think about it like an application that gets downloaded on your phone and it's intuitive," Kos says. "But when you actually put it in play, it does take change management.”

To manage this, Baptist Health engages in "application rationalization," a process of auditing their tech stack to eliminate redundancy, unless that redundancy serves a specific resiliency purpose, such as maintaining cash flow if a primary system fails.

The Bottom Line

Ultimately, Kos views these technological shifts through the lens of operational efficiency and financial return. Whether the ROI comes from headcount reduction, resource shifting, or simply keeping up with volume to improve patient experience, the goal is sustainability.

By bridging the gap between IT technicality and revenue cycle strategy, Kos and Baptist Health demonstrate how health systems can approach AI with skepticism of hype, a reliance on structured governance, and a focus on collaboration.