Franklin, Jacob; Chu, Ling; McDonald, Samuel A.; Turer, Robert W.; Webb, Janet; Steitz, Bryan D.; Yi, Haoyang; Chen, Qingxia; McCoy, Allison B.; Nall, Carolynn; Chen, Catherine; Rousseau, Justin F.; Lehmann, Christoph U.; Willett, Duwayne L.; Kumah-Crystal, Yaa A.; Mize, Dara E. (2026).听.听Discover Artificial Intelligence, 6(1), 550.听
Ambient AI scribes听are designed to reduce the time clinicians spend documenting patient visits by automatically generating clinical notes during appointments. To better understand how these tools perform in real-world practice, this study evaluated听Dragon Copilot听across two academic medical centers. A total of 120 physicians and advanced practice providers completed surveys assessing the system’s usability, documentation quality, and overall satisfaction. Overall, clinicians rated Dragon Copilot as highly usable and reported that it fit well into their clinical workflows. Many participants said the tool reduced the time spent on documentation, improved their ability to focus on patients during visits, and enhanced the overall patient encounter. However, the AI scribe did not eliminate broader challenges such as burnout, frustration with electronic health records, work-life balance, or cognitive workload. Overall satisfaction with the tool was positive but mixed, indicating that experiences varied among users. As one of the first multi-site evaluations of an ambient AI scribe, the findings suggest that Dragon Copilot can improve documentation workflows and support clinician well-being, while also highlighting that it is not a complete solution to the broader challenges clinicians face in everyday practice.
