Start by evaluating existing clinical workflows
The health systems seeing the strongest AI outcomes share a common discipline: They evaluate underlying workflows before layering technology on top.
AI can dramatically accelerate a well-designed process, helping to surface insights faster, reduce documentation burden, and free clinicians to focus on patient care. But no tool, however sophisticated, can compensate for a workflow that wasn't serving clinicians in the first place. Wolters Kluwer’s Shadow AI survey found 57% of healthcare providers and administrators had encountered an unauthorized AI tool in the workplace. Of those who reported using them, 50% did so for a faster workflow, and 1 in 3 cited either a lack of approved tools or that the approved tools lacked the desired functionality.
Clinicians turning to AI tools means leaders need to ask some key foundational questions before any rollout:
- Is this the best system in place to scale?
- Should they consider workflow improvements before layering in AI efficiencies?
- Where are some quick wins that can deliver AI value in the short-term?
Build governance that moves at the speed of innovation
Governance has become one of the most consequential factors in successfully scaling AI, partnering with clinical teams, and pivoting as business needs change.
A strong governance framework includes clear criteria for tool evaluation, explicit policies on approved and unapproved use, and regular review cycles that keep pace with the technology landscape. When clinicians have access to approved, high-quality tools that genuinely meet their needs, the incentive to experiment with unauthorized alternatives diminishes. Governance done well doesn't restrict innovation—it focuses it and empowers care teams.
Embedding those policies within EHR workflows, rather than relying solely on enterprise communications, helps ensure they're accessible for clinicians and professionals and improves policy awareness. This communication is essential—the Future Ready Healthcare Report found 44% of clinicians weren’t aware of AI policies in their organization, and 29% weren’t sure if they existed.
Governance can also include streamlining evaluation pathways, upskilling clinical teams on AI literacy, and creating structured channels that connect frontline needs directly to senior decision-makers to continue adjusting policy as innovation and workflow needs evolve.
