Clinicians' rationale for editing ambient AI-drafted clinical notes: persistent challenges and implications for improvement

Clinical Snapshot

Key Findings

• P Value: Not applicable — qualitative study

• Effect Size: Not applicable — qualitative study; no quantitative effect sizes reported

• Primary Outcome: Three primary motivations for editing AI-drafted clinical notes were identified: (1) improving clinical accuracy and specialty-specific precision; (2) reducing medico-legal and liability risk; and (3) meeting billing, coding, and documentation standards

• Nnt Or Sensitivity: Not applicable — qualitative study; five categories of editing triggers identified: transcription errors, speaker attribution mistakes, overconfident statements without evidence, missing key clinical details, and AI's lack of patient contextual information

• Confidence Interval: Not applicable — qualitative study

Clinical Application

Findings are immediately actionable at multiple levels: clinicians can adopt adaptive communication strategies during consultations; institutions can develop structured review guidance and training programmes; vendors can prioritise transcription accuracy, specialty customisation, and EHR integration improvements. Implementation requires no additional resources beyond existing AI tool deployment infrastructure. Ambient AI clinical documentation tools are emerging in Australian healthcare, with growing interest from hospital networks and general practice. The Australian Digital Health Agency's National Digital Health Strategy and the RACGP's digital health frameworks are relevant governance contexts. The TGA has not yet established a specific regulatory pathway for ambient AI documentation tools, though they may fall under Software as a Medical Device (SaMD) provisions. Medico-legal concerns identified in this study are directly applicable to the Australian context, where the Medical Board of Australia holds clinicians responsible for the accuracy of signed clinical records regardless of how they were generated. PBS billing and MBS item number documentation requirements add an additional layer of coding accuracy concern analogous to the US billing findings. RACGP standards for medical records (Standards for general practices, 5th edition) require that records be accurate, contemporaneous, and clinician-verified — making the editing and review behaviours described in this study directly relevant to Australian GP and specialist practice. Clinicians in outpatient settings using ambient AI documentation tools; findings are most directly applicable to academic health system contexts but have broader relevance to any clinician reviewing and signing AI-generated clinical notes

Abstract

OBJECTIVE: The use of ambient AI documentation tools is rapidly growing in US hospitals and clinics. Such tools generate the first draft of clinical notes from scribed patient-provider conversations, which clinicians can then review and edit before signing into electronic health records (EHR). Understanding how and why clinicians make modifications to AI-generated drafts is critical to improving AI design and clinical efficiency, yet it has been under-studied. This study aims to address this gap. MATERIALS AND METHODS: We conducted semistructured interviews with 30 clinicians from the University of California, Irvine Health who used a commercial ambient AI tool in routine outpatient care. We invited them to describe how and why they edited AI drafts based on both their personal experience and review of some real-world examples identified from our previous studies. RESULTS: Modifications to AI drafts were primarily made to improve clinical accuracy and specialty-specific precision, reduce medico-legal and liability risk, and meet billing, coding, and documentation standards. Such editing was necessary due to reasons such as transcription errors, speaker attribution mistakes, overconfident statements without evidence, missing key clinical details, and AI's lack of information about the patient context. CONCLUSION AND DISCUSSION: Improving ambient AI documentation will require coordinated effort from vendors, institutions, and clinicians. Key targets include core model reliability (eg, transcription accuracy), specialty- and encounter-level customization, clinician-level personalization, more effective EHR integration, and institutional support (eg, training, governance, and standardized review guidance), complemented by clinicians' adaptive communication strategies that strengthen human-AI collaboration.

References

1. 1.Guo, Y., Hu, D., Yang, Z., Chow, E., Tam, S., Perret, D., Pandita, D., & Zheng, K. (2026). Clinicians' rationale for editing ambient AI-drafted clinical notes: persistent challenges and implications for improvement. Journal of the American Medical Informatics Association. https://doi.org/10.1093/jamia/ocaf075