Across healthcare, organizations are investing heavily in AI to improve operational efficiency, support clinical decision-making, and create better experiences for patients and members. As adoption accelerates, the conversation is becoming less about what AI can do and more about what it takes to deploy it responsibly.
That question shaped a recent executive roundtable I had the privilege of moderating at the Emids Healthcare Summit 2026 in Nashville. Bringing together leaders from across provider organizations, health plans, health technology companies, and consulting firms, the discussion explored what it will take to build trust at scale as AI becomes part of everyday healthcare operations.
While each organization is at a different stage of its AI journey, several common themes emerged.
Trust Must Be Earned
Trust begins long before an AI solution reaches production. It is built through transparency, explainability, and confidence that recommendations can be understood and validated.
As AI becomes embedded in clinical and operational workflows, organizations are asking more sophisticated questions about how decisions are made, how recommendations can be verified, and how accountability is maintained. Explainability is now fundamental to user confidence and organizational adoption.
One perspective that resonated throughout the discussion was that healthcare often holds AI to a higher standard than people, and rightly so, given how often AI produces confidently wrong results. Human judgment naturally allows for nuance and uncertainty, while AI systems are expected to consistently explain and prove their reasoning before organizations are willing to rely on them in high-impact environments.
Ultimately, trust is not created by the technology itself. It is earned through the confidence people have in how that technology performs.
Building Trust Requires More Than Regulation
As AI adoption accelerates, organizations are navigating an increasingly complex governance landscape.
Many are balancing expectations from regulators, customers, legal teams, security leaders, and compliance officers, often evaluating the same AI solution through multiple frameworks. While emerging standards are beginning to provide direction, many organizations continue to develop their own governance models as they determine how best to evaluate AI risk.
One recurring observation was that organizations are increasingly learning from one another. Conversations about AI are extending beyond technology teams to include legal, compliance, privacy, and security leaders, with many looking to peers for practical approaches to governance and accountability.
This collaborative approach reflects an industry that recognizes trust cannot be created in isolation. It will require shared learning and evolving best practices.
Governance Should Reflect Risk
Not every AI application carries the same level of responsibility, and governance models should recognize those differences.
Administrative automation, software development, and workflow optimization present different considerations than AI supporting clinical decisions or patient care. Applying the same governance model to every use case can slow innovation in lower-risk areas and leave higher-stakes applications without enough oversight.
An important distinction surfaced during the discussion. Governance should be proportionate to the level of discretion an AI system has over business or clinical outcomes.
Human oversight also emerged as an important mechanism for building confidence. Rather than slowing innovation, keeping experts in the loop allows organizations to validate outputs, learn from real-world use, and expand AI adoption with greater assurance.
From AI Experimentation to Enterprise Value
The discussion also reflected a noticeable shift in mindset. One phrase captured it particularly well: healthcare is moving from the “fear of missing out” to the “fear of messing up.”
The excitement surrounding AI remains, but organizations are becoming more disciplined about where they deploy it, how they govern it, and how they measure success. Rather than pursuing AI simply because new capabilities continue to emerge, leaders are increasingly focused on solving meaningful business problems and demonstrating measurable value.
Another perspective that surfaced was the importance of being a “fast follower.” For many organizations, learning from proven implementations and validated outcomes is a more sustainable path than attempting to be first with every innovation.
The conversation suggested that enterprise AI success will be measured less by the number of pilots launched and more by the ability to deliver meaningful improvements across operations, clinical workflows, customer experiences, and business performance.
Trust at Scale Is a Continuous Journey
Perhaps the strongest takeaway from the discussion was that organizations build trust throughout the AI lifecycle.
Governance does not end with implementation. As models evolve, regulations mature, and new use cases emerge, organizations will need continuous monitoring, ongoing validation, and regular reassessment of how AI performs in real-world settings.
Healthcare has always balanced innovation with responsibility. AI should be no different.
As organizations move from experimentation to enterprise adoption, the real challenge is deploying AI in ways that clinicians, employees, regulators, and patients can trust.
That, ultimately, is what operationalizing AI responsibly looks like.
See more insights from the Emids Healthcare Summit 2026.
→ https://www.emids.com/summit/
Special thanks to the executive leaders who participated in the Trust at Scale roundtable discussion and generously shared their perspectives, experiences, and lessons learned. Their valuable contributions significantly enriched the insights presented in this article.
Roundtable Participants:
Clark Golestani, Board Director, CSensei Group
Nathan Kennedy, VP, Deputy CISO, Molina Healthcare
Martin Graf, Senior Partner, Oliver Wyman
Sashi Kodali, VP & CMIO, HCA Healthcare
LuAna Boykins, Transformation Officer, Triple-S Salud
Enda Murphy, CTO & CISO, IVX Health
Chiranjiv Singh, VP & GM, Pixel, Tempus
Jordan Taggart, SVP, Payer & PBM Growth, Vida Health
