From Frustration to Relationship

The "Dr. Google" of yesterday is the "Dr. AI" of today. If you've had a patient arrive with a stack of printouts from ChatGPT or Gemini, you know the feeling. It can be frustrating to spend precious appointment time debunking misinformation when you need to be focusing on real-time care.

From a patient's perspective, they feel their agency has been lost when their research and interest in their own health is quickly dismissed. I've heard a medical professional who dismissed their patient's research with a simple, "Did you go to medical school to even understand what you are asking? Ask you're question once you do." Sadly, we all can think of a doctor who would give this type of response.

But what if we see this differently? This trend isn't just about a new source of questionable advice; it's a sign of heightened patient engagement. Patients are curious, proactive, and trying to understand their health using the tools available to them.

Instead of waiting to react, we can lead.

By proactively establishing a framework for discussion, we can guide this curiosity and reinforce our role as their trusted expert. Imagine initiating the conversation about AI before they even ask. You could establish expectations by covering three simple points:

How is AI used within your medical ecosystem? Briefly, explain if and how AI tools (like those for reading scans or analyzing data) are used in your hospital.

How does it help with patient care? Clarify how these tools support, but don't replace your clinical judgment in making decisions about their care.

How do you think patients should use AI? Offer your professional guidance on how patients can use AI tools safely, perhaps for summarizing information or drafting questions for their next visit.

This simple, proactive approach transforms the dynamic from a potential debate into a trusting partnership. It channels patient curiosity productively, reduces clinician frustration, and builds a stronger therapeutic alliance.

Let's turn a point of friction into an opportunity for connection. I'd love to chat about this with you if this is something you and/or your healthcare setting is looking to explore.

About Dan Noyes

Dan Noyes operates at the intersection of healthcare AI strategy and governance. After 25 years leading digital marketing strategy, he is transitioning his expertise to healthcare AI, driven by his experience as a chronic care patient and his commitment to ensuring AI serves all patients equitably. Dan holds AI certifications from Stanford, Wharton, and Google Cloud, grounding his strategic insights in comprehensive knowledge of AI governance frameworks, bias detection methodologies, and responsible AI principles. His work focuses on helping healthcare organizations implement AI systems that meet both regulatory requirements and ethical obligations—building governance structures that enable innovation while protecting patient safety and advancing health equity

Want help implementing responsible AI in your organization? Learn more about strategic advisory services at Viable Health AI

This Isn't a Handful of Patients — It's One in Three

Let me put a number on the "stack of printouts" problem I described above. Rock Health's 11th annual Consumer Adoption of Digital Health Survey, which polled 8,000 U.S. adults in December 2025, found that 32% of Americans have used an AI chatbot for health information — double the 16% who said the same just a year earlier Rock Health Consumer Adoption of Digital Health Survey. KFF's Tracking Poll on Health Information and Trust, fielded around the same time, landed on an identical figure: about a third of adults turning to AI for health information and advice, putting AI chatbots roughly on par with social media as a health information source KFF Tracking Poll on Health Information and Trust. West Health and Gallup put the figure at 1 in 4 Americans, or more than 66 million people West Health-Gallup survey on AI use for healthcare.

These aren't casual queries. Rock Health found 59% of AI health users searched for treatment options after a diagnosis, 56% asked AI to help figure out what's causing their symptoms, and 41% used AI to interpret lab or imaging results Rock Health and Microsoft data on AI chatbot health use. Pew Research's newest report, published in August 2026, confirms the pattern: a quarter of Americans use AI chatbots specifically to diagnose symptoms, and similar shares use them to make sense of information they already got from a provider, like understanding a diagnosis or decoding lab results Pew Research: Why Americans Use AI Chatbots for Health. That last category should stop every clinician cold. Patients aren't only using AI instead of us. A meaningful share are using it to double-check us — after the visit, on their own time, often at 11pm with no one to ask a follow-up question.

The Trust Data Cuts Both Ways

Here's the part that should reframe how you think about the "Did you go to medical school" reflex I mentioned earlier: patients don't fully trust their health systems to manage AI responsibly either. A JAMA Network Open analysis found that 65.8% of respondents reported low trust in their health care system to use AI responsibly, and 57.7% reported low trust that their system would ensure an AI tool wouldn't harm them Patients' Trust in Health Systems to Use Artificial Intelligence. Meanwhile, a JMIR study found that simply mentioning a doctor uses AI to assist in diagnosis measurably decreased patients' trust in that doctor and reduced their stated intention to seek help Impact of AI-Assisted Diagnosis on American Patients' Trust. Put those two findings together and you get the real strategic problem: patients are using AI on their own in large numbers, yet they trust neither an AI-disclosing doctor nor their own health system's AI governance by default.

This is exactly why the three-question framework I laid out — how AI is used in your ecosystem, how it supports rather than replaces your judgment, and how patients should use it themselves — isn't just good bedside manner. It's a direct response to a documented trust gap, and disclosing proactively performs better than disclosing reactively. A JAMIA Open survey found patients overwhelmingly want transparency (92%) and clinician oversight (82%) before they'll trust an AI-generated recommendation at all JAMIA Open survey on patient trust in clinical AI. Notably, that same study found roughly 64% of patients said they would trust AI-generated recommendations given to their doctor for things like sepsis detection or medication adjustment — trust in the tool, channeled through the clinician, is achievable. It just has to be earned with disclosure, not assumed.

Health Literacy Changes the Conversation You Should Have

One nuance the three-point framework doesn't address directly: not every patient needs the same conversation. Research on AI chatbot design and trust found that a patient's health literacy significantly moderates how they respond to AI-mediated information, with effects on both cognitive and emotional trust differing by literacy level Bridging the health literacy gap through AI chatbot design. Separately, a TechTarget-reported survey found 54% of patients say it's at least somewhat difficult to distinguish accurate health information from inaccurate information when guidance conflicts, and 56% said they'd already encountered conflicting health information from different sources Patients unsure what to trust amid health information overload. If more than half your patients are already navigating conflicting information before they ever open a chatbot, your "how should you use AI" talking point needs to include a concrete filter, not just a general endorsement of caution.

There's an equity dimension too, and it should matter to any hospital leader thinking about population health. KFF found that one in five adults using AI for health information cited difficulty accessing or affording care as a reason, with larger shares among younger and lower-income patients KFF: One in three adults have used AI chatbots for health advice. For a meaningful slice of your patient population, the chatbot on their phone isn't competing with a $40 copay visit — it's filling a gap where the visit wasn't affordable or accessible in the first place. That reframes the AI-literate patient sitting across from you not as a nuisance, but as someone signaling an access problem your health system can actually do something about.

• Lead with the three-point framework at intake or annual visits, before a patient shows up with printouts — proactive disclosure measurably outperforms reactive correction.
• Ask what specific AI tool a patient used (general-purpose chatbots dominate at roughly three-quarters of AI health users, versus single digits for provider- or payer-built tools) so you know what kind of information they were actually given.
• Treat repeated AI-assisted questions about affordability or access as a signal, not a symptom of misinformation — a fifth of AI health users are turning to it because care felt out of reach.
• Calibrate literacy-level: match your explanation of AI's role to the patient's demonstrated health literacy rather than a one-size script.

Continue the Conversation

If this resonated, here is where to go next: Join the AI-in-Healthcare Workshop · Get the Books · Contact Dan