Want Dr. Calendine’s articles to stand out in your Google and AI search results? Add him as a preferred source (one tap).
AI in healthcare is no longer a prediction; it is quietly woven into the visit you had last month. A new national survey from Healio, published in May 2026, asked 618 health care professionals across more than 35 specialties how they actually use artificial intelligence at work. The answers surprised even those of us who follow this closely.
I want to walk you through what the survey found, what it means for you as a patient, and how I think about these tools in my own orthopaedic practice here in Franklin, Tennessee.
How Many Doctors Are Using AI in Healthcare Right Now?
About 72 percent of health care professionals now use AI tools regularly or occasionally in their work, according to the Healio survey. That includes doctors and nurses in private practices, academic centers, and community hospitals alike.
Here is the finding that deserves your attention: only 11 percent of them ever got formal training on AI from their workplace. Nearly 8 in 10 taught themselves, learning by trial and error. And more than 40 percent of respondents whose organizations do not formally support AI reported using it anyway.
After more than twenty years in medicine, I have watched new technology arrive in waves: electronic records, digital imaging, and robotic surgical platforms. The pattern with AI is different. Adoption is happening person by person, faster than hospitals can write the rulebook.
What Do Doctors Actually Use AI in Healthcare For?
Most AI use in medicine today happens behind the scenes, on paperwork rather than patients. The survey found the top uses were documentation and note generation (55 percent of organizations), clinical decision support (32 percent), and research or literature review (27 percent).
Documentation and the fight against burnout
The biggest single use is the least dramatic one: writing notes. AI scribes can listen to a visit and draft the record, which frees the doctor to look at you instead of a screen. In the survey, 73 percent of clinicians who use AI said it has had a positive impact on their work.
That matters for patients too. A doctor who spends less of the evening typing is a doctor with more energy for the next day’s clinic.
Decision support, not decision making
Some tools help summarize research, flag drug interactions, or highlight findings on an X-ray for a physician to review. Notice the wording: support. In every legitimate use I know of, a trained clinician reviews the output and makes the call.
The survey reflects that caution. Clinicians ranked proven clinical accuracy as the top requirement for trusting any AI tool, well ahead of cost or speed.
Why Is There a Trust Gap in AI in Healthcare?
Doctors use AI far more than they trust it, and honestly, that is the right instinct. Only 18 percent of surveyed professionals reported strong trust in AI for clinical decision making. Their top worries were accuracy errors, called “hallucinations” (70 percent), legal liability (59 percent), and leaning on the technology too much (58 percent).
The survey also found comfort drops as AI gets closer to the patient. About 69 percent of clinicians are comfortable using AI behind the scenes, but only 36 percent are comfortable using it directly in front of patients.
Here is what I tell my patients: that gap is not a failure. It is professional skepticism doing its job. Medicine has always adopted new tools this way; we test, we verify, we keep the human in charge.
Is AI in healthcare safe for patients?
AI is safe when it works under a qualified clinician’s supervision, and risky when it replaces one. The tools in use today draft, suggest, and summarize; they do not diagnose you on their own or decide your treatment. Your protection is the physician who reviews every output, plus growing oversight from regulators and professional societies.
That is also why the survey’s optimism is paired with a demand for guardrails. More than 90 percent of respondents said clinicians should play a central role in designing and selecting AI tools, and 83 percent expect AI to be an essential part of health care within 5 years.
The Training Gap: Most Doctors Learned AI on Their Own
The most fixable problem in the survey is education. Only 49 percent of health care professionals have had any AI education at all.
Just 11 percent got structured training from their institution, and fewer than 5 percent hold a formal certification. Only 5 percent are very satisfied with the training resources available to them.
I am an advocate for AI in medicine, and that is exactly why this bothers me. A powerful tool without training is how good technology gets a bad reputation. Surgeons do not pick up a new implant system or a new surgical platform and simply wing it; we train, we proctor, we measure results.
AI deserves the same discipline. Hospitals and practices that invest in real education, clear policies, and honest measurement will serve their patients better than those that let every clinician improvise.
How I Use Technology in My Own Practice
In my practice, technology has one job: making care more precise and more personal, never less. I perform more than 700 hip and knee replacements each year, many with the Mako® robotic-arm assisted system. It builds a 3D plan of your anatomy from a CT scan, then helps me place an implant with a precision that supports my surgical judgment rather than replacing it.
Transparency note: I serve as a consultant for Stryker, the company behind that platform.
That experience shapes how I look at AI. The robot never operates on its own; I am in control of every step. AI should be held to the same standard everywhere in medicine: a skilled tool in trained hands, with the physician accountable for the outcome.
It also shapes what I ask before adopting anything new: Is there evidence it works? Do I understand how it reaches its answers? Does it make the patient experience better?
If a tool cannot pass those three questions, it does not belong in patient care.
What Should Patients in Middle Tennessee Ask Their Doctor?
You have every right to ask how AI touches your care, and a good physician will welcome the question. Whether you see me in Franklin or a doctor in Nashville, Brentwood, Columbia, or anywhere in Middle Tennessee, these questions will get a useful conversation started:
- Do you use AI tools in my care, and for what tasks?
- Does a clinician review everything the AI produces?
- How is my health information protected when these tools are used?
- Has your team received training on the tools you use?
If your doctor uses an AI scribe during your visit, you should be told, and you can decline. In my experience, patients are far more comfortable with these tools once someone simply explains them. The survey backs that up: trust rises with transparency.
The Bottom Line on AI in Healthcare
AI in healthcare has moved from experimental to everyday, and the question is no longer whether doctors will use it but how responsibly they will use it. The Healio findings show a profession that is enthusiastic and careful at the same time: 71 percent support AI in health care today, while insisting on accuracy, oversight, and training before trusting it with clinical decisions.
I share both the enthusiasm and the caution. Used well, AI gives physicians more time and better information. Used carelessly, it erodes the trust that medicine runs on.
The difference comes down to education, clear rules, and honest conversation with patients. That is the standard I hold in my own practice.
If you are considering hip or knee replacement and want a surgeon who embraces proven technology while keeping the human relationship at the center of care, I would be glad to see you. You can schedule a consultation online or call the Bone and Joint Institute of Tennessee at (615) 791-2630. We are located at 3000 Edward Curd Lane, Franklin, TN 37067.
This article is for educational purposes only and does not replace professional medical advice. Always consult a qualified healthcare provider, such as a board-certified orthopaedic surgeon, before making medical decisions. Dr. Cory Calendine is a consultant for Stryker.
References
- Healio. The AI Trust Gap in Health Care: Perspectives on Adoption, Trust and Clinical AI. Healio Special Report. May 2026.
- American Medical Association. Augmented Intelligence in Medicine: AMA Principles and Policy. AMA; updated 2025.
- U.S. Food and Drug Administration. Artificial Intelligence and Machine Learning (AI/ML)-Enabled Medical Devices. FDA.gov; updated 2025.
Watch: Dr. Calendine reacts to the new AI survey data
