Ambient AI Scribes Capture Only What Is Spoken, Perspective Warns

Ambient AI scribes are being deployed across health systems to reduce documentation burden and clinician burnout, with encouraging early results. But a new Perspective in the Journal of General Internal Medicine argues that these tools have a blind spot that goes beyond workflow: because they generate clinical notes from speech, they capture only what is said during an encounter, not what clinicians observe without verbalizing or judge without articulating.

Author Henry Bair contends that this omission is not merely a technical limitation but an epistemological one. Drawing on the dual-process model of clinical reasoning, research on clinician intuition, and the verbal overshadowing literature from cognitive psychology, the Perspective suggests that audio-based documentation may create a structural incentive toward verbalization that risks interfering with tacit, nonanalytical expert judgment.

The downstream consequences identified include directional distortion of the clinical record, reduced detection of premature diagnostic closure, and the potential erosion of trainees’ appreciation for preverbal expertise. In other words, a note built from the spoken record may systematically underrepresent the rapid, experience-based judgments that experienced clinicians make without putting them into words.

The Perspective proposes two responses. Health systems should evaluate ambient scribes for reasoning fidelity rather than transcription accuracy alone, and medical educators should teach the distinction between tacit judgment and bias, so that learners understand what these tools can and cannot capture.

Bair’s conclusion is measured rather than alarmist: ambient scribes solve a real problem, but documentation systems built around what can be heard may gradually devalue clinical reasoning that cannot be spoken. As health systems continue scaling these tools, the argument suggests that the most important question is not how accurately a scribe transcribes the conversation, but whether the resulting record faithfully reflects the full scope of clinical thinking.

Subject of Research: Epistemological limitations of ambient AI scribes in clinical documentation

Article Title: What Ambient AI Scribes Cannot Hear

Article References: Bair, H. (2026). What Ambient AI Scribes Cannot Hear. Journal of General Internal Medicine. https://doi.org/10.1007/s11606-026-10779-8

Image Credits: AI Generated

DOI: 10.1007/s11606-026-10779-8

Keywords: ambient AI scribes, clinical documentation, clinician burnout, clinical reasoning, dual-process model, tacit knowledge, verbal overshadowing, diagnostic closure, medical education, health systems, Ambient, Scribes