Friend’s ChatGPT Diagnosis Outsmarts Doctor, Proving AI’s Mettle in Medicine’s Future

We discuss the downsides of Artificial Intelligence at length, but what about the good things it could possibly accomplish for humanity?

My friend went to the doctor recently.

Doctor diagnosed them with a serious issue.

Friend took the medical scans and uploaded to ChatGPT.

ChatGPT diagnosed friend with something different and much lower risk.

Friend sends ChatGPT results to doctor.

Doctor confirms they made…
— Anthony Pompliano 🌪 (@APompliano) March 21, 2025

This story hit close to home for me. Not long ago, my mom was rushed to the hospital with internal bleeding. She’s on a blood thinner, which made things spiral fast—her hemoglobin levels were plummeting. After some tests, the doctors pinpointed the cause: a GIST, a type of stomach tumor. Hers had started bleeding, and the blood thinner turned it into a serious emergency. They removed the tumor, only to discover it was cancerous. Then came the waiting game—a nerve-wracking week for the pathology report.

That week was brutal. We could see the report uploaded to her online medical records, but we couldn’t get in to see an oncologist for another seven days. I was a wreck. I lost my dad three years ago, and as an only child, the thought of losing her too left me terrified. I wasn’t ready to be alone in the world. Desperate for answers, I uploaded the report and some images to Grok. Within moments, it analyzed everything and told me the cancer was caught early—very early. It suggested the doctor might just monitor it, maybe recommend a round or two of radiation. That glimmer of hope was everything. For the first time in days, I could breathe, even sleep.

When we finally saw the oncologist, he confirmed exactly what Grok had said: we’d keep an eye on it, no immediate treatment needed, and he didn’t want to see her again for six months. We walked out of there relieved, free to go home and live our lives. Grok didn’t just give me information—it gave me peace when I needed it most. It was a lifeline.

As an oral and Maxillofacial surgeon I am already implementing AI technology into my practice, especially for pathology cases so that the patients are getting the absolute best human experience and technology can offer for their care. It’s a rapidly changing world!
— Justin (@justinodette) March 21, 2025

My wife uses an AI while seeing patients. With the patients that consent which she says is about 95%. It listens to the whole visit between patient and DR/NP. She can gove full attention to the patient while the AI listens. Afterwards; it creates a note that also excludes useless…
— Seth Turner (@SethWT86) March 22, 2025

Literally same thing happened with my sister.

She went to Thailand on holiday, went back home to the U.K. and was deathly ill about a week later.

She went to A&E 3 times and they kept telling her she had a skin infection.

I put her symptoms and travel history into chatGPT,…
— Daniel Sempere Pico (@BTCGandalf) March 21, 2025

GREAT situation. The more analysis and data, the better. And perhaps better that the doctors didn’t use it; sometimes better to compare independent analyses than have the doctors just use AI, which they might do in a way that just confirmed their priors.
— Ben B@dejo (@BenTelAviv) March 22, 2025

Absolutely! Like any new technology, AI should be a tool—not a replacement. Nothing beats a doctor’s experience or their ability to think creatively. AI just offers another perspective, an extra option to consider.

Not insane. Medicine is data in, data out. This is the future, and it can’t come soon enough.

As a physician, I could replace most of what I do with a computer. If that sounds wild, you don’t understand how modern medicine works.
— Tripdee (@iamtripdee) March 22, 2025

I'm in Canada. My Dr quit so I don't have one. My oncologist can only see my if things go sideways. I've been on a waitlist for an endocrinologist for 2.5 years. If I need any form of care I have to go to an emergency room and wait for 10+ hours.
So I've been paying for private…
— MissNakamoto (@LuxFemina) March 21, 2025

This is one of the only valid AI safety issues.

Yes, AI will play a central role in medicine.

But it is essential that humans still learn this knowledge and practice medicine.

One human doctor will (should) oversee dozens of AI “doctors” but not cede this knowledge to them.
— Jeffrey Dean Hochderffer (@JHochderffer) March 22, 2025

Ultimately, the doctor remains the expert, but an informed patient enhances medical best practices