Healthcare organizations are moving aggressively toward AI adoption. Still, the pressure to innovate quickly is colliding with the realities of governance, compliance, interoperability, and patient outcomes. Unlike many industries, healthcare cannot afford uncontrolled experimentation. The margin for error is simply too high.

In the second part of this two-part series, Tom MacDougall, former Chief Information and Technology Officer at L.A. Care Health Plan, speaks with Robert Lutton, Vice President at Sandhill Consultants, about responsible AI adoption, the growing importance of interoperability, and why governance must become foundational to modern healthcare operations.

Part 1 covered what it takes to transform data culture, why real-time clinical data is replacing the old claims-driven model, and how AI can be deployed responsibly in healthcare.

AI adoption in healthcare remains early-stage

As healthcare leaders face increasing pressure to adopt AI quickly, MacDougall says organizations must resist the urge to scale too fast: “It boils down to a walk, crawl, run type methodology.”

Rather than deploying AI broadly across the enterprise, organizations should begin with tightly controlled pilots, validate outcomes, and expand gradually.

MacDougall notes that some organizations are also using multiple models to cross-check outputs and reduce operational risk.

The conversation reflects a broader reality across healthcare: many organizations are still in the experimentation phase and are working to establish operational trust before scaling AI more aggressively.

Analytics is helping health plans improve care delivery

MacDougall also discusses how healthcare organizations are using analytics to better understand provider performance, member behavior, and treatment outcomes: “We collect a lot of data in terms of cost and use analysis.”

One focus area is identifying whether members are accessing the appropriate level of care at the right time, particularly around emergency room utilization.

By benchmarking provider performance and analyzing treatment outcomes, organizations can identify care gaps and improve decision-making across the network.

“We take a look at what care they are getting from their providers and benchmark that care back against other providers.”

While controlling costs remains important, MacDougall emphasizes that the larger objective is improving member outcomes.

Interoperability and governance will shape the future

Looking ahead, MacDougall believes healthcare organizations must prioritize governance and interoperability if they want AI and modern analytics initiatives to succeed: “AI governance will be key for rolling it into their operating model.”

He also points to standardized healthcare data as one of the industry’s most important long-term priorities.

According to MacDougall, trusted and portable data creates enormous operational value across healthcare ecosystems: “You can take it and move it around the enterprise at any given point, knowing that it’s accurate.”

He adds that lineage and traceability will become increasingly critical as AI systems become more embedded in healthcare operations.

CDO Magazine appreciates Tom MacDougall for sharing his insights with our global community.