Hacker News new | ask | show | jobs
by xiphias2 192 days ago
Another poorly written article that doesn't even specify the LLM being used.

Both ChatGPT o3 and 5.1 Pro models helped me a lot diagnosing illnesses with the right queries. I am using lots of queries with different context / context length for medical queries as they are very serious.

Also they have better answer if I am using medical language as they retrieve answers from higher quality articles.

I still went to doctors and got more information from them.

Also I do blood tests and MRI before going to doctors and the great doctors actually like that I go there prepared but still open to their diagnosis.

2 comments

The lack of awareness to advocate using LLMs for medical advice on an article about how bad of an idea that is... It's astounding.
There are people receiving life-saving medical advice from LLMs every single day.

The problem isn't getting medical advice from LLMs, it's blindly trusting the medical advice a LLM gives you.

You do not need to trust the LLM for it to be able to save your life, you do need to trust the LLM for it to be able to harm you.

>There are people receiving life-saving medical advice from LLMs every single day.

Is this just a vibe thing or has anyone actually done some statistics on this?

Sounds like it's something you assume is true solely based on the fact that you personally feel like LLMs are good at medical advice.

>Is this just a vibe thing or has anyone actually done some statistics on this?

There's extensive reporting on this from pretty decent news sources, and countless forum posts from seemingly real non-throwaway accounts which aren't otherwise spamming LLM content.

> based on the fact that you personally feel like LLMs are good at medical advice.

No, I don't think that's the case. LLMs sometimes give good advice, the bad advice isn't harmful unless you blindly trust the advice.

People go to a doctor or two, their symptoms are dismissed as [insert very common problem here], ask LLM, receive advice suggesting that [uncommon condition] also fits the symptoms and then take that to a doctor.

This also happens using Google every day. LLMs aren't special, they just make searching the internet a bit easier.

Read to the bottom. I didn't specify the LLM because it doesn't matter. It's not the fault of the LLM, it's the fault of the user
It’s the fault of the tool!