Live data from Hacker News

I misused LLMs to diagnose myself and ended up bedridden for a week

blog.shortround.space

61–70 of 150 posts

Re: I misused LLMs to diagnose myself and ended up bedridden for a week

#61
post #37

In the UK we have the 111 NHS non-emergency telephone service - they don’t give medical advice but triage you based on your symptoms. Either a doctor will call you back, they will tell you to go the non-urgent care centre or A&E (ER) immediately.

In the EU we have 116117 which is not (yet?) implemented in all countries. It's part of the "harmonised service of social value" which uses 116 as a prefix and has also other helpline numbers like hotlines for missing children or emotional support.

Re: I misused LLMs to diagnose myself and ended up bedridden for a week

#62

Earlier quoted context omitted.

> a child's fractured arm ... Um what? The only way to diagnose a fractured arm is an xray. You can suspect the arm is fractured (rotating it a few directions) but ultimately a muscle injury will feel identical to a fracture especially for a kid. Please, if you suspect a fracture just take your kid to the doctor. Don't waste your time asking ChatGPT if this might be a fracture. This just feels beyond silly to me imag…

[flagged]

> How does this line up with your religious belief that doctors are infallible and should be 100% trusted?

Because the way you phrased it with the article in question made it sound like you hadn't first gone to the doctor. This isn't a question about doctors being fallible or not but rather what first instincts are when medical issues arise.

> uneducated comment filled with weird assumptions.

No, not uneducated nor were these assumptions weird as other commenters obviously made the same ones I did.

I'll not delete my comment, why should I? The advice is still completely valid. Go to the doctor first, not GPT.

Re: I misused LLMs to diagnose myself and ended up bedridden for a week

#63

Yikes. But I have to say that prompt is crazy bad. AI is VERY good at using your prompt as the basis for the response, if you say "I don't think it's an emergency" AI will write a response that is "it's not an emergency" I did a test with the first prompt and the immediate answer I got was "this looks like lyme disease".

Well I tested the first prompt on ChatGPT and Llama and Claude and not one of them suggested Lyme disease. Goes to show how much these piece of shit clankers are good for.

Llama said "syphilis" with 100% confidence, ChatGPT suggested several different random diseases, and Claude at least had the decency to respond "go to a fucking doctor, what are you stupid?", thereby proving to have more sense than many humans in this thread.

It's not a matter of bad prompting, it's a matter of this being an autocomplete with no notion of ground truth and RLHF'd to be a sycophant!

Just 100B more parameters bro, I swear, and we will replace doctors.

Re: I misused LLMs to diagnose myself and ended up bedridden for a week

#64
post #45

@shortrounddev2 can you please post the response ChatGPT gave in response to your prompt? That seems pertinent.

It'd be interesting to see the entire chat, lots of people seem to just keep using the same chat window and end up poisoning the LLM with massive amounts of unhelpful context.

Re: I misused LLMs to diagnose myself and ended up bedridden for a week

#65

Earlier quoted context omitted.

No, the lesson here is never use an LLM to diagnose you, full stop. See a real doctor. Do not make the same mistake as me

"Don't ask LLMs leading questions" is a perfectly valid lesson here too. If you're going to ask an LLM for a medical diagnosis, you should at the very least know how to use LLMs properly. I'm certainly not suggesting that you should ask LLM for medical diagnoses, but still, someone who actually understands the tool they're using would likely not have ended up in your situation.

If you're going to ask an LLM for a medical diagnosis, stop what you're doing and ask a doctor instead. There is no good advice downstream of the decision to ask an LLM for a medical diagnosis

Re: I misused LLMs to diagnose myself and ended up bedridden for a week

#66
They didn't say what model they used. The difference between GPT 3.5 and GPT 4 is night and day. This is exactly what I'd expect from 3.5, but 4 wouldn't make this mistake.

Note: I haven't updated this comment template recently, so the versions may be a bit outdated.

Re: I misused LLMs to diagnose myself and ended up bedridden for a week

#67

Earlier quoted context omitted.

> a child's fractured arm ... Um what? The only way to diagnose a fractured arm is an xray. You can suspect the arm is fractured (rotating it a few directions) but ultimately a muscle injury will feel identical to a fracture especially for a kid. Please, if you suspect a fracture just take your kid to the doctor. Don't waste your time asking ChatGPT if this might be a fracture. This just feels beyond silly to me imag…

[flagged]

>How does this line up with your religious belief that doctors are infallible and should be 100% trusted?

Hilariously, this is the second time you posted this exact line, to yet another person who _didn't say this_!

Re: I misused LLMs to diagnose myself and ended up bedridden for a week

#68
Interesting story. I want to agree with the general advice not to use it for that - especially if that is how you use it. And I want to preface this with: Don’t take this as advice, I just want to share my experience here. I tend to do it anyway and had fairly large success so far but I use the LLM differently if I have a health issue that bothers me. First I open Gemini, Claude and ChatGPT in their latest, highest thinking budget installment. Then I tell them about my symptoms I give a fairly detailed description of my person and my medical history. I prompt them specifically to ask detailed questions like a physician would and ask them to ask me to perform tests to rule out or zoom in on different hypothesis about what is might have. After going back and forth, if they all agree on a similar thing or a set of similar things I usually take this as a good sign I might be on the right track and check if I should talk to a professional or not (edging on the side of caution). If they can’t agree I would usually try to get an appointment to see a professional and try to get sooner rather than later if anything potentially dangerous popped up during the back and forth or if I feel sufficiently bad.

Now, I live in Germany where in the last 20 years our healthcare system has fallen victim to neoliberal capitalism and since I am publicly insured by choice I often have to wait for weeks to see a specialist so more often than not LLMs have helped me stay calm and help myself as best as I can. However I still view the output less as a the output or a medical professional and try to stay skeptic along the way. I feel like the augment my guesswork and judgement, but not replace it.

Re: I misused LLMs to diagnose myself and ended up bedridden for a week

#69
post #15

Earlier quoted context omitted.

If the LLM starts telling me it can't answer my question, I switch LLMs

And when you’ve finally switched to the last LLM and it tells you it can’t answer you, and you decide not to use LLMs for important life-changing questions anymore, your world will be a better place.

Not really, no. I know I can't trust what the LLM tells me, I still want the answer.

This should be a configurable option.

Re: I misused LLMs to diagnose myself and ended up bedridden for a week

#70

They wrote... "Turns out it was Lyme disease (yes, the real one, not the fake one) and it (nearly) progressed to meningitis" What does "not the fake one" mean, I must be missing something?

Disclaimer: not a doctor (obviously), ask someone who is qualified, but this is what the ID doctor told me:

Lyme is a bacterial infection, and can be cured with antibiotics. Once the bacteria is gone, you no longer have Lyme disease.

However, there is a lot of misinformation about Lyme online. Some people think Lyme is a chronic, incurable disease, which they call "chronic lyme". Often, when a celebrity tells people they have lyme disease, this is what they mean. Chronic lyme is not a real thing - it is a diagnosis given to wealthy people by unqualified conmen or unscrupulous doctors in response to vague, hard to pin symptoms

Post reply on HN