The health care industry is still unlocking the full potential of artificial intelligence to boost efficiency and improve care while reducing costs. At the same time, however, so-called shadow AI, in which different AI-powered tools and chatbots are used without proper approval processes, poses risks and challenges well beyond the IT department.
"In 2025, shadow AI surged across health care organizations as staff across all aspects of care sought ways to improve efficiency amid persistent burnout, staffing shortages and other factors," said Alex Tyrell, senior vice president and chief technology officer at Wolters Kluwer.
In many cases, adoption and innovation are outpacing policy and enterprise decision-making, leading employees to use any tool they can get their hands on to accomplish their tasks. When this happens, leadership loses the ability to regulate or have full security oversight of tools within the organization, and clinical leaders have greater concerns about variations in care resulting from disparate tools.
Wolters Kluwer Health recently surveyed more than 500 full-time providers and administrators about their perceived use of AI tools, risk tolerance and shadow usage. Four in 10 had encountered an unauthorized AI tool in their organization but did not use it, while an additional 17% admitted using an unauthorized AI tool. The survey highlighted several key gaps between health system administrators and providers in tool usage, risk tolerance and perceptions of policy communications.
- Administrators are more optimistic about AI's impact. When asked if they believe AI will significantly improve health care within the next five years, administrators expressed more optimism than providers. The largest split was 48% of administrators and 34% of providers selecting "strongly agree" as a response. Sixteen percent of providers were "neutral," compared to just 6% of administrators
- Clinicians are more likely to experiment with unsanctioned tools/ Although burnout and administrative burden have been lessening, they still remain at untenable levels. Clinicians are looking for the most efficient ways to support patients and get clinical answers, which often includes using free AI tools that could jeopardize outcomes and security, and possibly introduce patient risk.
- Administrators are more likely than providers to believe AI policies are strongly communicated. A final gap identified in the survey is how AI policies are created and communicated. Nearly one-third of administrators, including those with clinical backgrounds, indicated they were involved in reviewing, developing or updating AI policies, while only 9% of providers were.
"As shadow AI continues to be more prevalent, clinicians should only use purpose-built GenAI systems that are trained on expert-validated evidence, transparent with source citations and capable of tailored recommendations," said Greg Samios, CEO of Wolters Kluwer Health. "GenAI will provide an increase in staff efficiency and care quality, but we must preserve safety and clinician-patient relationships by reframing workflows that elevate GenAI from a tool to a partner, keeping patients at the center of care."
© Arc, All Rights Reserved. Request academic re-use from www.copyright.com. All other uses, submit a request to TMSalesOperations@arc-network.com. For more information visit Asset & Logo Licensing.