Peer-reviewed research

Ambient Artificial Intelligence Scribes: A Pilot Survey of Perspectives on the Utility and Documentation Burden in Palliative Medicine

James Patterson et al., Healthcare (Basel), 2025 · DOI 10.3390/healthcare13172118

Read on October 17, 2026

What it is: A prospective quality improvement pilot at an academic palliative medicine service in Halifax, Nova Scotia, in which two residents used an ambient AI scribe and also dictated the same visits, with documentation time compared for each resident. It was not randomized and had 17 cases with both times recorded.
How much weight it carries: Peer-reviewed and open access, but small and not randomized, with two residents, short duration and self-assessed note ratings. The authors state it is insufficiently powered to draw conclusions about AI scribe benefits in real-world use.

Facts taken from this source

The authors of the palliative medicine pilot state that, because of its small sample size and short study duration, the study is insufficiently powered to draw conclusions about AI scribe benefits in real-world contexts.
“Given the small sample size and short study duration, this study is insufficiently powered to draw conclusions with respect to AI scribe benefits in real-world contexts.” (Discussion)
In a small palliative medicine pilot, documentation time was significantly lower with an ambient AI scribe than with dictation for one of two residents, and showed no significant difference for the other.
“For Resident 1, the mean documentation time was significantly lower for the AI scribe compared with dictation ... However, for Resident 2, there was no significant difference in the documentation time between the AI scribe and dictation” (Results, Documentation time (AI scribe vs. dictation))