AI Scribes Won't Fix What's Actually Breaking Physicians
- Heath Jolliff, DO
- Jul 8
- 5 min read
The burnout data is moving in the right direction. Here's why that's not enough.
I want to start by saying something that might surprise you coming from a physician coach: the AI scribe research is real, and it matters.
We are not talking about vendor hype. A 2025 study published in JAMA Network Open found a 21.2% absolute reduction in the prevalence of burnout among physicians who used ambient documentation technology for 84 days, along with a 30.7% improvement in documentation-related well-being.
A separate JAMA Network Open study found clinicians using AI scribes spent 20.4% less time on notes per appointment and logged 30% less after-hours EHR work.
That's the "pajama time" that has quietly been stealing years from physicians' lives. A peer-reviewed NEJM Catalyst analysis of Kaiser Permanente's program reported that ambient AI saved physicians an estimated 15,791 hours of documentation time across 2.5 million patient encounters.
These are not small numbers. This technology is doing something meaningful, and physicians deserve to know it.
Now let me tell you what the same research reveals, and why it should concern every health system leader in the country.
The Number Nobody Is Talking About
After the best-performing AI scribe study, conducted with enthusiastic early adopters at one of the world's most sophisticated academic health systems, 30.7% of physicians were still burned out.
Let that sit for a moment.
The intervention worked. Burnout dropped significantly. And nearly one in three physicians still crossed the finish line of an 84-day pilot program depleted, detached, and at risk of leaving. The study authors noted this themselves: given the 22-30% survey response rates, the findings "likely represent the experience of more enthusiastic users." Broader rollouts often show more modest results. Indeed, a large-scale study published in JAMA on April 28, 2026 found that while AI scribes provided some relief, the time savings were often modest and the impact on overall documentation burden was inconsistent across diverse clinical settings.
This is not a knock on the technology. It is an honest reading of what the technology can and cannot do.
AI scribes address documentation burden. Documentation burden is one branch of a tree with deep, tangled roots. And we have been so focused on the branch that we have nearly forgotten the roots exist.
What Is Actually Breaking Physicians
Here is what I know from working with physicians inside healthcare systems: the ones who are quietly planning their exits are not, as a rule, drowning in notes. They are drowning in something harder to name, and harder to fix with software.
The research calls it moral distress. I call it the experience of being asked, day after day, to practice in ways that conflict with why you went into medicine.
A March 2026 national survey study published in JAMA Network Open, drawing on responses from 5,741 physicians, found that 41.6% reported high levels of moral distress, compared with just 14.2% of non-physician workers. More striking: physicians with high moral distress scores were 2.5 times more likely to experience burnout than those with low scores (75.1% vs. 30.7%). And 34.4% of physicians with high moral distress reported plans to leave their current position within 24 months, nearly double the rate of their less distressed colleagues.
This is not burnout from charting too many notes. This is burnout from being trapped in systems that prioritize throughput over patient care, that erode clinical autonomy under the weight of administrative mandates, and that leave physicians feeling like billing units rather than healers. An AI scribe does not touch any of that. It cannot.
The drivers the research consistently identifies go well beyond the EHR:
Loss of professional autonomy and control over clinical decisions
Values misalignment with institutional leadership
Understaffed teams and inadequate support structures
Lack of meaningful input into scheduling and workload
A slow erosion of the sense of purpose that drew physicians to medicine in the first place
These are organizational and relational problems. They require organizational and relational solutions.
Why This Is Not Just a Wellness Issue
If the human cost is not enough to command attention, the financial case should be.
A landmark study by Han, Shanafelt, Sinsky and colleagues published in the Annals of Internal Medicine estimated that physician burnout costs the U.S. healthcare system approximately $4.6 billion annually, with individual replacement costs running between $500,000 and over $1 million per departing physician. Burnout-driven turnover among primary care physicians alone costs payers an estimated $979 million every year. And a 2026 study by Dhruv Khullar published in JAMA Internal Medicine found that burned-out family physicians are nearly 1.5 times more likely to change practices or exit clinical care altogether.
Health systems are not investing in AI scribes out of generosity. They are investing because the math on physician turnover is catastrophic. But if the underlying drivers of moral distress and leadership dysfunction remain unaddressed, the technology buys time without solving the problem. Physicians get their evenings back. They still do not feel like their institution sees them. The exit interview still comes. It just takes a little longer.
What Actually Has to Change
The literature on sustainable physician wellbeing consistently points to the same interventions: restoring autonomy, rebuilding meaning, developing leadership at the unit level, and creating cultures where physicians can raise concerns without fear of reprisal.
None of that comes in a software subscription.
This does not mean AI scribes are not worth adopting; they are. Think of them as removing a genuine obstacle. But removing an obstacle is not the same as building a road. Physicians who are burned out, morally injured, or quietly disengaging need something more than recovered administrative time. They need support in reconnecting with who they are as clinicians and leaders. They need systems that are structured to actually retain them, not just to make them slightly less miserable.
Sustainable recovery from burnout, at the individual and organizational level, requires the kind of honest, structured work that does not happen in a 15-minute wellness webinar. It requires examining what meaning looks like now, what leadership capacity exists at the team level, and what structural changes would make it possible to practice with integrity again.
A Word to Health System Leaders
If you are a CMO, department chair, or physician leader reading this, investing in AI scribes is worth it. Make it. But do not let it substitute for the harder conversation about what your culture is actually asking of your physicians.
The physicians who are one patient complaint away from resigning are not suffering from a documentation problem. They are suffering from a systemic problem that has been rebranded as a personal one. Giving them cleaner notes does not change that equation.
The real question is not, "How do we make burnout more manageable?" It is "What would it take to build a system physicians do not want to leave?"
That is a different question entirely, and it is one worth asking before your next physician exits to concierge medicine, a locums arrangement, or medicine altogether.
All statistics in this article are drawn from peer-reviewed research. If you'd like the full reference list, drop a comment below or send me a message; I'm happy to share it.
If you're a physician or physician leader navigating burnout, moral injury, or a system that no longer feels sustainable, I'd welcome a conversation.
This is the work I do.
Heath Jolliff, DO, ACC
Executive Physician Coach | Founder, Physician Coaching Solutions
Three decades in clinical medicine taught me what breaks physicians. The last six years in executive coaching have taught me what rebuilds them. I work with physicians who are burned out, at a crossroads, or practicing in systems that no longer reflect why they went into medicine.
Comments