Related Experiment Videos
Propagation of Interpreter Errors by Ambient AI Scribes: Study Using Simulated Clinical Encounters
Alexandra Rabotin1,2, Efren Aguilar3, Saitiel Sandoval Gonzalez3
1Family Medicine Residency Program, Mission Community Hospital, 14860 Roscoe Blvd, Panorama City, CA, 91402, United States, 1 818-787-2222.
JMIR Medical Informatics
|July 28, 2026
Summary
Ambient AI scribes copied interpreter mistakes into clinical notes during simulated English and Spanish encounters. Further research is needed to assess AI scribe accuracy in diverse clinical settings.
Area of Science:
- Medical Informatics
- Clinical Linguistics
- Artificial Intelligence in Healthcare
Background:
- Interpreter errors can impact patient safety and care quality in multilingual clinical settings.
- Ambient AI scribes are increasingly used to document patient encounters, but their performance with interpreters is not well understood.
Purpose of the Study:
- To evaluate how ambient AI scribes propagate interpreter errors into clinical notes.
- To identify patterns in error propagation based on speaker role and error type.
Main Methods:
- Simulated clinical encounters in English and Spanish were conducted.
- Ambient AI scribes were used to generate clinical notes from these encounters.
- The generated notes were analyzed for the presence and patterns of propagated interpreter errors.
Main Results:
- Ambient AI scribes propagated interpreter errors into clinical notes.
- The patterns of error propagation varied depending on the speaker's role (e.g., clinician, patient, interpreter).
- Specific error types were more likely to be propagated than others.
Conclusions:
- Ambient AI scribes may inadvertently introduce or amplify errors in interpreter-mediated clinical care.
- The findings underscore the need for rigorous evaluation of AI scribe performance in multilingual healthcare contexts.
- Further development and validation are required to ensure the accuracy and reliability of AI scribes in diverse clinical scenarios.