Editorial Implications of Artificial Intelligence–Powered Ambient Scribes Pragya Kakani, PhD; Austin S. Kilaru, MD, MSHP; Melinda B. Buntin, PhD Investment in artificial intelligence (AI)–powered ambient scribes has surged, as health systems across the country pilot this technology.1,2 Recent studies have found that ambient scribes reduced documentation burden, with associated enthusiasm among clinicians regarding their potential.3-6 However, a new Viewpoint by Nong and Neprash7 in JAMA Health Forum raises concern that AI scribes may increase health care spending by increasing diagnostic and procedure coding intensity and recommending low-value services. Nong and Neprash raise important points that have not been addressed empirically and are among several unanswered questions that may be important for policy. Understanding the full implications of ambient scribes requires rigorous evaluation, not only of documentation burden and patient- and clinician-reported satisfaction, but also of downstream impacts, both positive and negative, on health system outcomes. These include outcomes related to health care spending as well as quality of care and equity. Health Care Spending As argued by Nong and Neprash,7 ambient scribe companies are incentivized to provide a financial return on investment to their customers: health systems and physician practices. Although not the only means to offer a financial return, there is a real risk that financial return will be generated through more intensive diagnostic coding. Prior work has convincingly shown that health systems strategically upcode services when financially advantageous, contributing to growing health care costs.8 However, health care organizations do not upcode at equivalent rates, due in part to the cost and sophistication required.9 Ambient scribes have the potential to automate upcoding by capturing more data than clinicians might otherwise record and then optimizing documentation to maximize the complexity of recorded patient encounters. Ambient scribes may also increase health care utilization by using notes to predict additional care a patient might want or need and prompting clinicians or patients to seek that additional care. However, it remains uncertain to what extent ambient scribes will increase overall health care spending for multiple reasons. First, although the technology is evolving rapidly, to date there is limited evidence on the financial return on investment for health systems, with some reports suggesting minimal returns.2 Second, it is possible that any increases in spending due to upcoding may be counterbalanced by reductions in administrative spending empowered by the use of ambient scribes, such as reductions in time spent by physicians and administrators in billing and prior authorizations.10 Finally, payers are also likely to respond strategically to coding trends. Their response may include using their own AI-powered tools to detect and respond to clinicians exhibiting upcoding, via claim denials, penalties, or network exclusion. Payers will need to be nimble in their responses to avoid cost escalations. Public payers, such as traditional Medicare, may be at a disadvantage, given their limited existing utilization management tools, and potentially slower policy implementation timelines. Quality of Care Despite their rapid adoption, there are also important questions regarding the potential impact of ambient scribes on health care quality. Early evidence suggests that ambient scribes may improve + Related article Author affiliations and article information are listed at the end of this article. Open Access. This is an open access article distributed under the terms of the CC-BY License. JAMA Health Forum. 2026;7(1):e256150. doi:10.1001/jamahealthforum.2025.6150 (Reprinted) January 9, 2026 1/3 Downloaded from jamanetwork.com by guest on 01/21/2026 https://jama.jamanetwork.com/article.aspx?doi=10.1001/jamahealthforum.2025.5771&utm_campaign=articlePDF%26utm_medium=articlePDFlink%26utm_source=articlePDF%26utm_content=jamahealthforum.2025.6150 patient experience because clinicians can avoid documenting while evaluating patients.3-6 Yet there is very limited evidence on the impact of ambient scribes on patient outcomes. As Nong and Neprash7 acknowledge, there is some optimism that the deployment of ambient scribes can identify opportunities to deliver guideline-concordant care, recognizing the need for important screening tests like vaccinations or recommended imaging. Data collected and processed by the ambient scribe would then prompt clinicians or patients to pursue additional testing. If recommended services are high-value, this may improve population health. There is additional hope that data collected by ambient scribes may eventually assist with diagnosis, providing additional suggestions for diagnostic testing and potentially reducing the need for specialty referral or accelerating the time to diagnosis.11 Ambient scribes may also improve the quality of clinical notes, better supporting longitudinal care for individual patients and population health insights. However, there are also substantive concerns that ambient scribes may change clinical practice, if not adversely affect quality. First, ambient scribes may increase the overall volume of documentation in select cases because manual editing of notes can be time-consuming.5,12 Although clinicians may train ambient scribes to edit notes over time, the excess information recorded from clinical encounters represents a paradigm shift in the intentionally selective ways that clinical information is communicated. Second, ambient scribes may make mistakes, due to confabulations or misinterpretation of human communication.13 Third, overreliance on ambient scribe documentation may erode clinical accountability, as clinicians may become too reliant on AI-enabled decision support or other prompts that result from ambient recordings.14 Collectively, these concerns do not warrant abandoning this technology but should prompt clinicians and health care systems to anticipate and mitigate any adverse effects. Health Equity Ambient scribes may also have multidirectional implications for health equity; this is an essential area for future research. On the one hand, ambient scribes may reduce disparities in the quality of documentation across patient group that can result from clinician bias or limited language proficiency.15 At the same time, ambient scribes may themselves not yet be optimized equally for all populations and especially across languages, potentially resulting in poorer documentation for non-English speakers. Another important concern is that patient trust of ambient scribes may vary, and ambient scribes may be more likely to erode patient trust for patients from historically marginalized communities. In addition, patients with varying technological literacy may have differential understanding and acceptance of any risks to privacy or clinical outcomes. It is also unclear whether any increases in upcoding or health care utilization would reduce or exacerbate disparities in revenues between safety net and other clinicians or patient access.16 Conclusions Although ambient scribes hold promise for reducing documentation burden, their broader implications for health care spending, quality, and equity remain uncertain and complex. The potential for increased upcoding and low-value care, unknown effects for health care quality, and potential for exacerbation of health disparities underscores the need for careful surveillance as this technology spreads. It is imperative for policymakers, researchers, and health care system leaders to remain vigilant to ensure that these technologies achieve their goals to improve efficiency without causing more insidious impacts on the health care system. ARTICLE INFORMATION Published: January 9, 2026. doi:10.1001/jamahealthforum.2025.6150 JAMA Health Forum | Editorial Implications of Artificial Intelligence–Powered Ambient Scribes JAMA Health Forum. 2026;7(1):e256150. doi:10.1001/jamahealthforum.2025.6150 (Reprinted) January 9, 2026 2/3 Downloaded from jamanetwork.com by guest on 01/21/2026 https://jama.jamanetwork.com/article.aspx?doi=10.1001/jamahealthforum.2025.6150&utm_campaign=articlePDF%26utm_medium=articlePDFlink%26utm_source=articlePDF%26utm_content=jamahealthforum.2025.6150 Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2026 Kakani P et al. JAMA Health Forum. Corresponding Author: Pragya Kakani, PhD, Weill Cornell Medicine, 575 Lexington Ave, New York, NY 10022 (pka4006@med.cornell.edu). Author Affiliations: Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland (B. Buntin); Senior Associate Editor, JAMA Health Forum (B. Buntin); Editorial Fellow, JAMA Health Forum (Kakani, S. Kilaru); Hospital of the University of Pennsylvania, Philadelphia (S. Kilaru). Conflict of Interest Disclosures: Dr Buntin reported an advisory board role at Harvard Medical Faculty Practice and unpaid and personal fees from Peterson Health Technology Initiative outside the submitted work. No other disclosures were reported. Disclaimer: Dr Kakani is an Editorial Fellow and Dr Buntin is a Senior Associate Editor of JAMA Health Forum but they were not involved in any of the decisions regarding review of the manuscript or its acceptance. REFERENCES 1. Aguilar M. With ambience’s new mega-round, AI scribes have announced nearly $1 billion in funding this year. Stat. June 29, 2025. Accessed September 4, 2025. https://www.statnews.com/2025/07/29/ambience-healthcare- ai-scribe-new-fundraise 2. Adoption of artificial intelligence in health care delivery systems: early applications and impacts (March 2025). Peterson Health Technology Institute. Accessed October 22, 2025. https://phti.org/ai-adoption-early-applications- impacts/ 3. Rotenstein L, Melnick ER, Iannaccone C, et al. Virtual scribes and physician time spent on electronic health records. JAMA Netw Open. 2024;7(5):e2413140-e2413140. doi:10.1001/jamanetworkopen.2024.13140 4. Duggan MJ, Gervase J, Schoenbaum A, et al. Clinician experiences with ambient scribe technology to assist with documentation burden and efficiency. JAMA Netw Open. 2025;8(2):e2460637-e2460637. doi:10.1001/jamanetworkopen.2024.60637 5. Shah SJ, Crowell T, Jeong Y, et al. Physician perspectives on ambient AI scribes. JAMA Netw Open. 2025;8(3): e251904-e251904. doi:10.1001/jamanetworkopen.2025.1904 6. Angus DC, Khera R, Lieu T, et al; JAMA Summit on AI. AI, health, and health care today and tomorrow: the JAMA Summit report on artificial intelligence. JAMA. 2025;334(18):1650-1664. doi:10.1001/jama.2025.18490 7. Nong P, Neprash HT. Unintended consequences of using ambient artificial intelligence scribes for billing. JAMA Health Forum. 2025;7(1):e255771. doi:10.1001/jamahealthforum.2025.5771 8. Dafny LS. How do hospitals respond to price changes? Am Econ Rev. 2005;95(5):1525-1547. doi:10.1257/000282805775014236 9. Sacarny A. Adoption and learning across hospitals: The case of a revenue-generating practice. J Health Econ. 2018;60:142-164. doi:10.1016/j.jhealeco.2018.06.005 10. Sahni N, Stein G, Zemmel R, Cutler DM. The potential impact of artificial intelligence on healthcare spending. National Bureau of Economic Research. 2023. Accessed December 3, 2025. www-nber-org.ezproxy.med.cornell. edu/system/files/chapters/c14760/c14760.pdf 11. D’Adderio L, Bates DW. Transforming diagnosis through artificial intelligence. NPJ Digit Med. 2025;8(1):54. doi:10.1038/s41746-025-01460-1 12. Tierney AA, Gayre G, Hoberman B, et al. Ambient artificial intelligence scribes: learnings after 1 year and over 2.5 million uses. NEJM Catal. 2025;6(5):CAT-25. doi:10.1056/CAT.25.0040 13. Asgari E, Montaña-Brown N, Dubois M, et al. A framework to assess clinical safety and hallucination rates of LLMs for medical text summarisation. NPJ Digit Med. 2025;8(1):274. doi:10.1038/s41746-025-01670-7 14. Topaz M, Peltonen LM, Zhang Z. Beyond human ears: navigating the uncharted risks of AI scribes in clinical practice. NPJ Digit Med. 2025;8(1):569. doi:10.1038/s41746-025-01895-6 15. Hagiwara N, Penner LA, Gonzalez R, et al. Racial attitudes, physician-patient talk time ratio, and adherence in racially discordant medical interactions. Soc Sci Med. 2013;87:123-131. doi:10.1016/j.socscimed.2013.03.016 16. Dragan KL, Desai SM, Billings J, Glied SA, et al. Association of insurance mix and diagnostic coding practices in New York State hospitals. JAMA Health Forum. 2022;3(9):e222919-e222919. doi:10.1001/jamahealthforum.2022.2919 JAMA Health Forum | Editorial Implications of Artificial Intelligence–Powered Ambient Scribes JAMA Health Forum. 2026;7(1):e256150. doi:10.1001/jamahealthforum.2025.6150 (Reprinted) January 9, 2026 3/3 Downloaded from jamanetwork.com by guest on 01/21/2026 https://jamanetwork.com/pages/cc-by-license-permissions/?utm_campaign=articlePDF%26utm_medium=articlePDFlink%26utm_source=articlePDF%26utm_content=jamahealthforum.2025.6150 mailto:pka4006@med.cornell.edu https://www.statnews.com/2025/07/29/ambience-healthcare-ai-scribe-new-fundraise https://www.statnews.com/2025/07/29/ambience-healthcare-ai-scribe-new-fundraise https://phti.org/ai-adoption-early-applications-impacts/ https://phti.org/ai-adoption-early-applications-impacts/ https://jama.jamanetwork.com/article.aspx?doi=10.1001/jamanetworkopen.2024.13140&utm_campaign=articlePDF%26utm_medium=articlePDFlink%26utm_source=articlePDF%26utm_content=jamahealthforum.2025.6150 https://jama.jamanetwork.com/article.aspx?doi=10.1001/jamanetworkopen.2024.60637&utm_campaign=articlePDF%26utm_medium=articlePDFlink%26utm_source=articlePDF%26utm_content=jamahealthforum.2025.6150 https://jama.jamanetwork.com/article.aspx?doi=10.1001/jamanetworkopen.2025.1904&utm_campaign=articlePDF%26utm_medium=articlePDFlink%26utm_source=articlePDF%26utm_content=jamahealthforum.2025.6150 https://jama.jamanetwork.com/article.aspx?doi=10.1001/jama.2025.18490&utm_campaign=articlePDF%26utm_medium=articlePDFlink%26utm_source=articlePDF%26utm_content=jamahealthforum.2025.6150 https://jama.jamanetwork.com/article.aspx?doi=10.1001/jamahealthforum.2025.5771&utm_campaign=articlePDF%26utm_medium=articlePDFlink%26utm_source=articlePDF%26utm_content=jamahealthforum.2025.6150 https://dx.doi.org/10.1257/000282805775014236 https://dx.doi.org/10.1016/j.jhealeco.2018.06.005 https://www-nber-org.ezproxy.med.cornell.edu/system/files/chapters/c14760/c14760.pdf https://www-nber-org.ezproxy.med.cornell.edu/system/files/chapters/c14760/c14760.pdf https://dx.doi.org/10.1038/s41746-025-01460-1 https://dx.doi.org/10.1056/CAT.25.0040 https://dx.doi.org/10.1038/s41746-025-01670-7 https://dx.doi.org/10.1038/s41746-025-01895-6 https://dx.doi.org/10.1016/j.socscimed.2013.03.016 https://jama.jamanetwork.com/article.aspx?doi=10.1001/jamahealthforum.2022.2919&utm_campaign=articlePDF%26utm_medium=articlePDFlink%26utm_source=articlePDF%26utm_content=jamahealthforum.2025.6150