For the past couple of years, though, my numbers had been inching up, and I was frustrated that I couldn’t seem to do much about them. I requested a phone call from my doctor — surely, she had better advice than what she wrote — but she messaged back that if I wanted to discuss my results, I had to set up another appointment.
As both a physician and a patient, I found the experience startling. Not because ChatGPT dazzled me with its scientific knowledge, but because it behaved the way I wish modern medicine, and its practitioners, still would.
I had always assumed the “human side” of medicine was the part A.I. couldn’t touch. Sure, I know doctors are turning to A.I. to help them break bad news, since patients seem to find messages crafted by bots more empathetic than those written by doctors. But, in practice, what I thought really mattered was that a person was delivering that care.
The chatbot didn’t just spit back generic advice. It asked questions about my daily life and figured out what I could realistically change. It suggested a short walk immediately after eating, something I’d never taken seriously. When I inquired about doing a longer activity, it told me that would likely offer only marginal benefit. Its recommendations were manageable and easy to follow. [...]
As a doctor, I was a little embarrassed to be using ChatGPT. But every interaction with, say, OpenEvidence, a professional medical A.I. tool, felt cold and sterile. It referred to me as if I were a case report, not a person with preferences and habits. I realized what was winning me over about ChatGPT wasn’t its ability to sift through the latest studies, or diagnose my ailments; but its unwavering messages of empathy and encouragement, and its endless willingness to listen and its patience. It’s not human, but it can model some traits we value most in human interaction.
I followed ChatGPT’s advice, and when my blood work improved, ChatGPT affirmed my progress and urged me to keep going. I doubt I would have made those changes — much less stuck with them — without that sustained back-and-forth. I certainly hadn’t before.
It’s a grim fact of American medicine today that doctors can’t come close to a chatbot’s availability. And when the health care system can’t reliably offer time, attentiveness and compassion, patients will go searching for them somewhere else, even from a machine we assumed could never feel human. A.I. may not replace doctors, but it will change what patients expect from us. Doctors need to adapt.
Before I used a chatbot for my own health concerns, the thought of telling a patient to “ask ChatGPT” was inconceivable — or at least something I considered terrible care. Now I’m not so sure. In certain situations, A.I. offers something patients clearly need and medicine has trouble fulfilling.
The reality is, many patients are already consulting A.I. Doctors can keep fearing or condemning those interactions, or they can figure out how to support people using A.I. tools for their health care — cautiously, with clear guardrails. I would never tell patients to ask ChatGPT or Claude for a diagnosis, but perhaps I would suggest they use it to make sense of a new condition or keep up with routine screenings — or translate “diet and exercise” into steps that actually fit into their lives, as I did. At the same time, we need safeguards built into these systems to protect people from real harm from dangerous advice.
My experience with the chatbot has already shifted how I interact with patients in the E.R., with only minutes to piece together fragments of their circumstances. When a patient asks the same question repeatedly, I try to listen for what’s behind it. Maybe she’s not after more medical facts.
by Dr. Helen Ouyang, NY Times | Read more:
Image: María Medem
[ed. I had this exact experience a month or so ago. Asked for a full blood workup to see if there were any problems. Called back two weeks later for results. No answer. Waited another week and went to the clinic in person to make sure my first request hadn't somehow gotten lost in the bureaucracy (which happens, frequently). Except, this time I was smart enough to ask for a print-out of my lab results. Again, after receiving no response from my doctor, I took a picture of the results, uploaded them to Anthropic's Claude and asked it to interpret them. At first I got the standard disclaimer that it doesn't do diagnoses, but then I asked it to just interpret the results so I'd know what all the coding meant, the various ranges of acceptability etc., and it gave me a detailed response. Much better than I'd ever gotten from doctors before, who'd mostly just say (if they responded at all) "oh yeah, everything looks ok, some things look a little high, others ok". And that's it. No explanation or guidance on anything. Like follow-ups were a burden (that couldn't be billed for an office visit). I'll always use AI from now on to evaluate my results. Doctors (and hospitals) have brought this upon themselves.]
