Key Takeaways
- Ambient speech technology is the #1 clinical AI use case by a wide margin in healthcare organizations.
- Among Epic users, 62.6% have deployed ambient documentation tools as of June 2025, with Microsoft DAX Copilot (now Dragon Copilot) leading adoption.
- Physician burnout, often due to excessive time spent on documentation, is costly can be effectively addressed through use of AI scribes.
Walk into any exam room today, and you’ll see it: the doctor’s eyes aren’t on you, they’re on a screen. Physicians spend approximately 44% of their time on EHR documentation. This is true across multiple different types of practices and physicians, and it’s become unavoidable. Without proper documentation, there are no medical records that allow for reimbursement. Over the years, the time spent documenting has eroded into the doctor-patient relationship. As a result, there is more distrust of physicians by patients, and burnout and administrative overload for physicians.
In 2025, something shifted. A clinical artificial intelligence (AI) technology moved past pilots into widespread adoption by solving the problem physicians hate most: documentation. That technology is ambient AI, and it’s already deployed by 79% of healthcare organizations using AI — making it the #1 clinical AI tool in the U.S.
What Ambient AI Documentation Actually Does
Ambient AI documentation uses natural language processing (NLP) and generative AI to convert spoken clinician-patient conversations into structured clinical notes. The technology listens passively through a mobile app, captures the encounter in real time, and generates a draft note with history of present illness, review of systems, physical exam, assessment, and plan. The clinician reviews and edits, then signs off — in seconds, not minutes. More importantly, it occurs after the patient leaves, allowing the doctor to examine the patient instead of typing during the visit.
Leading solutions like Microsoft Dragon Copilot, Abridge, and Ambience Healthcare all work on the same principle: use AI to eliminate clerical burden, freeing clinicians to focus on care.
The Adoption Surge: From Pilot to Standard of Care
According to KLAS Research’s December 2025 Healthcare AI Update, 79% of healthcare organizations using any form of AI have deployed ambient speech technology. This is not a niche adoption story. KLAS validated this finding through interviews with 3,370 respondents across 1,742 unique organizations — and the data is unambiguous: ambient documentation is the dominant AI use case in healthcare, outpacing every other clinical and operational application.
Among hospitals using the Epic EHR system — which represents 42.4% of all U.S. hospitals — 62.6% had adopted an ambient AI tool as of June 2025. The three most commonly deployed solutions are Microsoft DAX Copilot (now integrated into Dragon Copilot), Abridge, and ThinkAndor, which together account for more than 80% of ambient AI implementations in Epic-using hospitals.
The adoption pattern is striking. Hospitals with higher patient workload, higher outpatient volume, and positive operating margins adopted ambient AI at significantly higher rates. Why? Because the return on investment (ROI) is measurable. Northwestern Medicine reported a 112% return on investment and a 3.4% service-level increase after deploying DAX Copilot.
This is an example of how we are no longer in the pilot-stage of technology anymore — this is production-scale deployment.
Why Clinicians Actually Use This Technology
Most clinical AI tools get buried in pilot purgatory — used by a handful of enthusiastic early adopters, then abandoned when the workflow friction becomes too high or the ROI too uncertain. Ambient documentation is different because it solves the problem clinicians hate most: documentation burden.
According to KLAS Arch Collaborative data from 2025, surveyed providers using ambient speech reported:
- 12% reduction in burnout related to staying after hours for documentation
- 10% reduction in physician burnout related to documentation
- Improved perceived EHR efficiency
These are not marginal improvements. The National Bureau of Economic Research found that the cost of burnout is approximately $80,000+ in lost productivity per physician.
The technology also addresses a core problem in healthcare staffing shortages. Clinician burnout is a leading driver of attrition, and documentation burden is consistently cited as a top contributor to burnout. By removing the single largest source of administrative frustration, ambient AI directly impacts physician retention.
Beyond the Note: Where Ambient AI Is Headed
Ambient documentation started as a solution to automate the clinical note. But the underlying technology — real-time capture and structuring of clinical conversations — is a platform that vendors are already expanding.
Microsoft’s Dragon Copilot for nurses, entering general availability in December 2025, extends ambient to nursing workflows — automatically generating flowsheet entries, nursing notes, and patient summaries.
Other emerging capabilities: order suggestions from patient conversations, one-click referral letters and after-visit summaries, clinical evidence summaries linked to transcripts, and billing code suggestions based on documented care. These move ambient AI from documentation automation to workflow orchestration.
Investment Implications for HealthTech
The ambient AI market is in early growth with a steep adoption curve. Key signals for investors:
- Enterprise EHR Integration: Deep integration with Epic or Oracle Health creates switching costs that protect market position.
- Expansion Beyond Physicians: Extending to nursing and allied health unlocks larger markets.
- Commoditization Risk: With 90+ vendors and free tiers emerging, price compression is inevitable. Winners will differentiate through workflow customization and specialty-specific models.
The Risks and What Could Go Wrong
Ambient AI is not without risks, and investors should be clear-eyed about where adoption could stall or reverse.
- Accuracy and Liability: If an ambient system generates a clinical note with factual errors and the clinician fails to catch it, who is liable? This remains legally murky. Early litigation around AI-generated documentation could dampen enthusiasm.
- Reimbursement Pressure: Payers could argue that if documentation is easier, physicians should see more patients for the same pay. If ambient AI becomes a productivity tool that shifts negotiating power to payers, physicians may resist adoption.
- Vendor Lock-In Backlash: Healthcare organizations are wary of becoming dependent on a single AI vendor, especially given the consolidation happening around Microsoft and Epic. If interoperability and data portability become pain points, organizations may push back.
- Workflow Disruption: Despite the promise of seamlessness, ambient AI still requires adoption support, training, and workflow redesign. Organizations that treat it as a simple technology swap — rather than a change management initiative — may see tepid adoption and minimal ROI.
These are manageable risks, but they are real. The difference between ambient AI succeeding at scale versus stalling out in the next 2-3 years will hinge on how vendors, health systems, and regulators address these friction points.
Conclusion
The doctor’s note as we’ve known it — the time-consuming, screen-facing task that has dominated clinical work for decades — is ending. What replaces it is a model where the conversation itself becomes the documentation, and the clinician’s attention stays on the patient. For the first time in healthcare AI’s history, we have a technology that is deployed at scale with measurable ROI and clinician buy-in.
For HealthTech investors, the signal is clear: ambient AI is not the future. It’s the present.
This article is intended for informational purposes only and does not constitute investment advice. Readers should perform their own due diligence and consult with a licensed financial advisor before making investment decisions.
Sanjana Vig MD, MBA
Dr. Vig is a co-founder and Chief Marketing Officer of Langar Holdings. She is a board-certified anesthesiologist specializing in Perioperative Management. She is also the founder The Female Professional, a website geared toward empowering professional women in life and their careers.
