The Hard Part of AI Is Not the Algorithm

Healthcare AI has made significant progress over the past decade, with models now performing strongly in controlled environments. Yet when these tools are introduced into real clinical settings, the results are often far less consistent. The gap between what works in theory and what works in practice has become increasingly visible.

Across health systems, a similar set of challenges continues to come up. Even high-performing tools can struggle to gain traction once they meet the realities of clinical workflows. Issues such as trust, alert fatigue, EHR integration, and governance processes all play a role. In many cases, ownership is unclear, reimbursement pathways are still evolving, and ongoing monitoring is difficult to maintain, which makes sustained adoption harder to achieve.

This roundtable will focus on what actually determines success once AI is deployed in real-world environments. Drawing on practical experience from those working inside health systems, the discussion will explore how organizations are approaching implementation, where things tend to break down, and what it takes to move from promising tools to something that is genuinely used and trusted.

Join us to discuss:

• Where AI initiatives encounter the most friction during real-world deployment
• How organizations are building clinician trust and encouraging adoption
• What governance, workflow, and ownership approaches are helping teams scale successfully