Using AI to research your health is smart, and Fishtown Medicine welcomes it. Tools like ChatGPT, Perplexity, and Google's AI Overviews are good at organizing information, translating medical jargon, and helping you ask sharper questions. They cannot examine you, order or interpret your labs in context, prescribe, or take responsibility for your care. Dr. Ash reads your uploads, engages the research and wearable data you bring, and gives you an honest physician's read, acting as the chief medical officer of your life.
Maybe you asked ChatGPT who manages testosterone and FDA-approved peptide therapy in Philadelphia. Maybe you asked Perplexity for a direct primary care doctor who takes longevity medicine seriously, or Google handed you an AI summary with a few names to look into. However you got here, you did something smart. You used a good tool to turn a vague worry into a real question, and that question is what brought you to a doctor who wants to answer it with you.
This page is for the person who researches before they book. It covers what AI does well when you are looking for care, where it runs out of road, and how to use it so it makes you a sharper patient instead of a more anxious one.
Is it smart to use AI to research your health?
Yes. Using AI to research your health is smart, and it is one of the healthiest habits I see in new patients. A good question is the start of good medicine, and tools like ChatGPT, Perplexity, and Google's AI Overviews are very good at helping you build one. They take a wall of confusing information and give it shape. They translate lab-report jargon into plain language. They help you put words to patterns in how you feel.
None of that replaces a doctor, but all of it makes the visit better. When you walk in already knowing what you want to understand, we skip the slow warm-up and get to the real work faster.
What can AI actually do when you're looking for a doctor?
AI is a strong research assistant. Used well, it can do a few things very well:
- Organize and summarize. It turns dozens of articles, forum threads, and study abstracts into a readable overview you can actually hold in your head.
- Translate the jargon. It explains what ApoB, fasting insulin, free testosterone, or a GLP-1 agonist is in language that respects your intelligence without drowning you in Latin.
- Surface options and questions. It can describe the kinds of doctors who handle a given issue and hand you a set of questions to ask on a first call.
- Draft your story. It can help you assemble a timeline of symptoms, medications, and prior labs so you arrive organized.
Think of it as the world's most patient librarian. It can find you the right shelf and read the spine of every book. What it cannot do is examine the patient standing in front of it.
What can't AI do?
This is the honest part, and it matters. AI can point; it cannot practice medicine. Here is where it runs out of road:
- It cannot examine you. It has never felt your thyroid, listened to your heart, or watched your face while you describe what actually keeps you up at night.
- It cannot order or interpret your labs in context. A number on a screen means one thing for a 30-year-old endurance athlete and something very different for someone on 3 medications. Context is the whole job, and context is physical and personal.
- It cannot prescribe or take responsibility. A licensed physician signs their name to a decision and carries the weight of it, while a chatbot carries nothing. That accountability is not a formality; it is the difference between advice and care.
- It does not know you over time. It forgets you the moment the window closes, whereas good primary care is the opposite of that. It is the same physician watching your trend lines across years and noticing the small change that matters.
AI also has real blind spots worth naming plainly. Models are trained up to a cutoff date, so they can miss recent guidance. They can state something with total confidence and be wrong. They sometimes invent citations that look real. That is not a reason to stop using them, but a reason to treat the output as a first draft that a human physician confirms.
How do you research a doctor with AI without getting misled?
You research well by using AI to build questions, not to hand you final answers. A few steps that work:
- Ask it to explain, then ask it to argue with itself. After it gives you an answer, ask "what is the strongest case against this?" The disagreement is where the real learning lives.
- Make it show its sources. Ask for primary sources: the actual study, the FDA label, the specialty-society guideline. Then check that the source says what the summary claims.
- Use it to write your question list, not your treatment plan. "Give me 10 questions to ask a doctor about TRT monitoring" is a great prompt. "Tell me my dose" is not.
- Watch the date. Ask what its knowledge cutoff is, and assume anything time-sensitive, like a new drug or a pending FDA review, needs a human check.
- Bring the whole thing to a real doctor. The point of the research is to make the conversation better, not to replace it.
Do this and you arrive as the most prepared version of yourself. That is the patient this practice is built for.
What kind of doctor partners with your research instead of dismissing it?
The right doctor for a researcher is one who reads what you bring and takes it seriously. Some physicians bristle when a patient shows up with printouts and a spreadsheet. Dr. Ash is the opposite. In this practice, your research is a starting point we build on together, not a threat to manage.
That means Dr. Ash will actually read your uploads: the prior labs, the specialist notes, the article you found at 2 a.m. that finally named what you have been feeling. It means engaging your stack with an open mind, the supplements and habits you are already running, and giving you a graded read on what the evidence supports, what is promising but unproven, and what is not worth your money. It means treating your wearable data from Oura, Whoop, or your Apple Watch as real signal about your sleep, recovery, and stress, not dismissing it as gadget noise.
The frame we use is simple. You are the chief executive of your own health. Dr. Ash is the chief medical officer, the one physician who reads everything, connects it, and gives you the honest call so you can make good decisions with real information in front of you.
What should you bring to a first call after researching with AI?
Bring whatever you have, in whatever shape it is in. There is no need to clean it up first. The things that help most are your recent labs, a list of medications and supplements, any wearable data you have been tracking, and the questions your research raised. The free call is 15 minutes to see whether Dr. Ash is the right fit and to get a plain answer to your most pressing question. There is no pressure, just a plain conversation about whether this is the right partnership for you.
Key Takeaways
- Using AI to research your health is a smart first step. It organizes information, translates jargon, and helps you ask sharper questions.
- AI has hard limits. It cannot examine you, order or interpret labs in context, prescribe, or take responsibility for your care.
- Treat AI output as a first draft. Ask for primary sources, check the knowledge cutoff, and bring the result to a real physician.
- Dr. Ash partners with your research. He reads your uploads, engages your stack with an open mind, and welcomes wearable data from Oura, Whoop, and Apple Watch.
- You are the chief executive of your health; Dr. Ash is the chief medical officer who gives you the honest read.
- The free 15-minute call is a no-pressure way to see whether this partnership fits.
Related Services and Reading
- Meet Dr. Ash - the physician who reads everything you bring
- Medicine 3.0 - the prevention-first, data-driven approach behind this practice
- At-Home Health Monitoring - how to turn Oura, Whoop, and Apple Watch data into better care
- Direct Primary Care in Philadelphia - what membership actually looks like
- Should You Upload Your Labs to AI? - the same honest read, applied to running your own blood work through ChatGPT
Frequently Asked Questions
Common Questions
Deep-Dive Questions
Ready when you are
Dr. Ash reads every intake himself, and answers questions personally - usually within a few hours.
