The patient on hold with your scheduling line already knows what she thinks is wrong.
Last night she talked it through with ChatGPT. Not a search box, a conversation. She described what she was feeling in her own words, answered a few follow-up questions, and came away with a plausible explanation and a short list of things to raise. It took a few minutes, and it felt like talking to someone who was listening.
Today she reaches your office and the experience runs in reverse. A clipboard, or a portal form that logs her out before she finishes. The same details she just spoke out loud, now re-entered by hand, in the order the form demands, next to the family history and the medication list she has typed a dozen times before. Half an hour of effort to hand over what she narrated in five minutes.
She feels the difference. One experience met her where she was. The other made her do the work twice.
Your patients stopped waiting
Here is what changed while the industry debated whether AI was ready for healthcare. Patients did not wait for the verdict.
In a single year, the share of Americans using AI chatbots for health information doubled. Sixteen percent in 2024. Thirty-two percent in 2025, according to Rock Health’s Consumer Adoption survey.
A jump like that does not settle back down. It becomes the baseline you plan around.
Roughly three in five adults have asked an AI tool a health or medical question in the last three months, OpenAI reported in December. People now consult it before a visit, after a visit, and sometimes instead of one, according to research from West Health and Gallup. More than half of the patients using AI for health are using it to check or make sense of symptoms.
This is not only the young or the tech-savvy. One of ChatGPT’s own ad campaigns shows an older woman talking to it like a neighbor. The interface has disappeared. What patients feel is that they can finally just say what is going on and get a straight answer.
The first conversation about a patient’s health is increasingly happening before your organization is in the room. You are not always the front door anymore. Sometimes you are the second opinion.
They adopted AI. They didn’t hand it their trust.
So the instinct is to rush. Bolt a chatbot onto the website, announce an AI initiative, catch up.
That instinct is about to cost you, because it misreads what patients are actually saying.
Most adults, including 56% of the people who already use AI, are not confident the health information it gives them is accurate, KFF found. Fewer than half, just 41%, say they would trust a health app or portal that used an AI chatbot to book their appointments or message their doctor. Seventy-seven percent are worried about the privacy of the medical details they type in.
Both things are true at once, and they do not cancel out. Patients lean on AI for convenience and still doubt it on accuracy. They will use it to get ready for a visit and refuse to let it make the call. That gap is not resistance for you to argue them out of. It reads more like instructions, telling you where AI belongs in their care and where it does not.
Your patients already drew the map
When you ask them, the line is sharp.
Patients are most comfortable with AI handling the parts of care that surround the visit. In a 2025 KLAS and Luma Health survey, 79% were comfortable with AI scheduling an appointment and 73% were fine with it handling notes and documentation. Ask about the clinical core and the number collapses. Only 37% were comfortable with AI involved in diagnosis.
That 37% is easy to misread as a wall around clinical AI. The truer read is about disclosure. When AI helps with a diagnosis, patients want to know a clinician is still making the decision, and they want to be told so plainly. Educate them on how much human judgment stays in the loop, and much of the discomfort drops.
The trade they will accept says the same thing. A majority, 57%, said yes to AI in the exam room, on the condition that it buys them more time and attention with their doctor, not less.
There is nothing timid about that. Patients are marking, with real precision, where AI helps them and where it needs to step back. Clear the path to care. Handle the remembering, the routing, the paperwork, the follow-up. Then leave room for the moment a person needs another person.
Your patients already decided where they want AI. They want it clearing the road to you, not standing in the road where you should be.
I walk through all of this on video this week, where AI earns its place in the patient experience and where it backfires: Your Patients Adopted Health AI Before You Did.
The question that actually matters
So the real question was never whether to adopt AI. It is where AI earns trust in the patient experience, and where it burns it.
Put it on the operational layer, the part around the clinical encounter, and you meet patients exactly where they already said yes. The scheduling that runs at 9pm when they finally have a minute. The reminder that actually reaches them. The intake that lets them say it once instead of typing it five times. The follow-up that arrives before they have to chase it. Every one of those is a place patients told a survey they welcome the help.
There is a financial reason to start there, not only a trust reason. When organizations actually track it, the fastest and largest returns from AI come from operational work like scheduling, documentation, and billing, well ahead of clinical uses (KPMG 2025 survey of healthcare organizations). Ambient documentation alone has handed clinicians back around four hours a week. On the revenue side, a missed appointment runs roughly $200, and organizations using AI to schedule and send reminders have reported cutting no-shows by a quarter to a third (industry analyses). Put AI where patients already trust it, and the return and the trust land in the same place.
The tools change fast. The rule underneath does not. Automate the predictable so your people have room for the moments only a person can hold. Technology should make the human moments more dependable, not stand in for them. The human never comes out of it.
What this means for you
For a health system, this is a leakage problem, not a technology one. Every patient who resolved their question with a chatbot at midnight, then could not reach you the next morning, has already found a faster way in than yours. Speed on the operational layer is how you stay their first call instead of their fallback.
For an independent or small practice, the same principle applies and the stakes are more personal. You cannot answer the phone at 11pm. Something you trust can, for the narrow set of tasks patients are glad to hand off, so your team’s attention lands where it is worth the most. The patient who felt met is the patient who rebooks.
There is an equity edge here too. Lower-income adults are more likely to turn to AI precisely because cost and access barriers make it hard to reach you. If getting to you is slow, AI becomes the substitute. If getting to you is fast and AI-assisted, you capture trust across the very populations most organizations keep losing.
So where AI goes is not guesswork. The survey data points to it directly. Put AI in the places patients said they welcome it, and keep it out of the places they said they do not.
Start here
Here is something you can do this week, and it costs nothing.
Pull the last week of calls to your front desk or scheduling line and sort them by what the patient actually wanted. You will see the same handful of intents over and over. Where do I park. What do I bring. Is my refill ready. Can I move my appointment. Those repeatable requests are where AI can carry the load. Hand them off, and your team is not just freed up for more minutes with patients. They are freed up to be present, which is the part no tool can do.
If you want to see where those gaps are quietly costing you, the Trust ROI Calculator walks you through it, free: patientexperiencestrategist.com/trust-tax.
Let’s get to work,
Ebony
Ebony Langston is the founder of The Patient Experience Strategist™ and a fractional Chief Experience Officer for healthcare organizations rebuilding patient trust as a margin strategy. With more than 20 years of operations experience supporting Fortune 100 healthcare organizations, she now works with leaders across the field, from health system C-suites and patient experience teams to independent and small-group practice owners. She writes weekly for the executives and clinicians turning patient experience from a cost line into a revenue engine.
Connect on LinkedIn | Subscribe to The Patient Experience Strategist