I have no doubt that a lot of garden-variety diagnoses and treatments can be done by an AI system that is fine-tuned and vetted to accomplish the task. I recently had to pay $93 to have a virtual session with a physician to get prescription for a cough syrup, which I already knew what to take before talking to her because I did some research/reading. Some may argue, "Doctors studied years in med school and you shouldn't trust Google more than them", but knowing human's fallibility and knowing that a lot of doctors do look things up on places like https://www.wolterskluwer.com/en/solutions/uptodate to refresh/reaffirm their knowledge, I'd argue that if we are willing to take the risk, why shouldn't we be allowed to take that risk on our own? Why do I have to pay $93 (on top of the cough syrup that costed ~$44) just so that the doctor can see me on Zoom for less than 5 mins and submit an order for the med?
With the healthcare prices increasing at the breakneck speed, I am sure AI will take more and more role in diagnosing and treating people's common illnesses, and hopefully (doubt it), the some of that savings will be transferred to the patients.
P.S. In contrast to the US system, in my home city (Rangoon, Burma/Myanmar), I have multiple clinics near my home and a couple of pharmacy within two bus stops distance. I can either go buy most of the medications I need from the pharmacy (without prescription) and take them on my own (why am I not allowed to take that risk?) OR I can go see a doctor at one of these clinics to confirm my diagnosis, pay him/her $10-$20 for the visit, and then head down to the pharmacy to buy the medication. Of course, some of the medications that include opioids will only be sold to me with the doctor's prescription, but a good number of other meds are available as long as I can afford them.
const_cast · 20m ago
The only reason this worked is because your situation was exceedingly simple.
The trouble is you are not educated enough to tell what is simple and what isn't. A cough could be a cough or it could be something more serious, only a "real" examination will reveal that. And sometimes even that's not enough, you need an examination by a specialists.
I'll tell you a story. Once upon a time I got pain in my balls. I went to a doctor and he felt around and he said he didn't feel anything. I went to another doctor and he felt something, but he had no idea what it was. He said could be a cyst, could be a swollen vein, could be an infection - he didn't even know if it was on the testicle or on the tube thingy.
Then I went to a Urologist. You can tell this man has felt up a lot of balls. He felt me up and said, "yup, that's a tumor" almost immediately. He was right, of course, and he ended up being the one to remove it too. Since I caught the cancer pretty early the chemotherapy wasn't too intense.
Point is, expertise matters when things aren't straight forward. Then, experience and perspective gets to shine.
imiric · 1h ago
> Why do I have to pay $93 (on top of the cough syrup that costed ~$44) just so that the doctor can see me on Zoom for less than 5 mins and submit an order for the med?
Because you're paying for the expertise of someone who studied for more than a decade which you won't get from a random web search.
An AI system with today's technology should be less trustworthy for medical diagnosis than a web search. At least with a web search you might stumble upon a site with content from experts, assuming you trust yourself to be able to discern expert advice from bot-generated and spam content. Even if a doctor is doing the searching instead of me, I would pay them only for their knowledge to make that discernment for me. Why you think an AI could do better than a human at that is beyond me.
Your question reminds me of that famous Henry Ford GE invoice story:
> Making chalk mark on generator: $1.
> Knowing where to make mark: $9,999.
hiatus · 49m ago
There's a limit though right? How about, if you can accurately predict the doctor's diagnosis it's free? If not you pay. This person needed a doctor's approval presumably for a prescription they couldn't get without authorization which leads to this gatekeeping. Not to mention also contributes to the insane medical costs in the US. $93 for 5 minutes is over 1k _an hour_ for what amounts to a rubber stamp (because how much can you really dig into a person's medical history and condition in 5 mins).
ChadNauseam · 58m ago
You answered why someone would want to pay $93, but not why they have to pay $93.
wubrr · 38m ago
I mean, it really depends on what you're trying to diagnose/do. Yeah, a math PhD might have spent a decade+ studying the subject. Doesn't mean I need to pay for their skills instead of using a calculator/program to solve my math problem.
> Why you think an AI could do better than a human at that is beyond me.
Why do you think an AI couldn't do better than a human, when we have ample evidence of computers/AI exceeding humans in many areas?
85392_school · 1h ago
AI systems have been improving. O3 now has the capability to decide to search multiple times as part of its response.
creata · 47m ago
Even o4-mini (free) uses web searches and runs Python scripts very eagerly. I'm not sure how long they'll be able to afford giving all of that away.
BeetleB · 2h ago
Where are you that you need a prescription to get cough medicine? The only ones I know of that require prescription are the ones with controlled substances.
HaZeust · 1h ago
Ain't nothing wrong with a little Wockhardt to get you through the day
dgfitz · 2h ago
> Of course, some of the medications that include opioids will only be sold to me with the doctor's prescription
onlyrealcuzzo · 1h ago
There are simply not enough doctors for doctors to be doing this work.
We have a massive, massive shortage of doctors.
The industry is doing everything they can to make it worse by the day, so I won't hold my breath that we'll get the slightest bit of respite.
It'd obviously be ideal if everyone could see a doctor for an affordable price any time they wanted.
We don't live in the ideal world.
This would be a HUGE win for most people.
nradov · 1h ago
This is why most routine primary care can be delivered by Nurse Practitioners and Physician Assistants working under a doctor's supervision.
closewith · 38m ago
NPs and community paramedics are definitely the future for a huge chunk of the current GP workload, but PAs are a failed experiment which are being rolled back pretty much everywhere.
nradov · 25m ago
Are you sure about that? The ranks of PAs have been expanding rapidly. I know there has been some political resistance over expanding their scope of practice but I'm not aware of it being significantly rolled back.
The US is the global outlier here and it's in the face of the evidence. The number of PAs globally is plummeting as they're abolished by national authorities.
ezst · 2h ago
I mean, if such is your typical experience with healthcare, what you need is a big shake up, don't count on AI to help you there! (but it might inflate the profit margins for the few running the show at your expense).
yapyap · 1h ago
You think you won’t have to pay the AI? lol.
hollerith · 1h ago
IMHO it is much easier to increase the supply of AI doctor hours than the supply of human doctor hours, and supply constraints are the main thing maintaining doctor hours at the high price it currently sells for.
imiric · 1h ago
Good lord. The idea that a system that produces pseudo-random output without any semantic understanding can be relied on to give accurate health-related information is deeply flawed and troubling. It's one thing to use these things for finding patterns in data, for entertainment purposes, and for producing nonsensical code a human has to fix, but entirely different to rely on them for health diagnosis or advice. I shudder at the thought that a medical practitioner I go to will parrot whatever an LLM told them.
This insanity needs to be regulated yesterday.
creata · 28m ago
Can you state more clearly why it's deeply flawed?
Because while LLMs obviously have massive limitations, so do humans, and it's not entirely clear to me that some synthesis of the two can't produce much better results than either on its own.
baby_souffle · 3m ago
> ... some synthesis of the two can't produce much better results than either on its own.
In theory, I agree with you. The world "some" is doing a lot of heavy lifting there though. I only hope that whatever definition of some emerges, it's not a horribly flawed one.
FlamingMoe · 1h ago
Good lord. The idea that a sleep‑deprived, bias‑prone mammal can be relied on to give accurate health‑related information is deeply flawed and troubling. It’s one thing to seek a human for empathy, small talk, or a routine vaccine, but entirely different to assume an overworked doctor will consistently out‑diagnose a modern data tool. I shudder at the thought that the practitioner I visit might simply repeat whatever a drug‑rep lunch pitch or outdated guideline left in their memory.
stavros · 1h ago
This insanity needs to be regulated yesterday.
imiric · 1h ago
Thankfully, it is. :)
hiatus · 43m ago
Are there regulations regarding doctor's working hours similar to say, truck drivers? I found this [1] article which states doctors are limited to 80 hours a week whereas truck drivers are limited to 70 hours in a 7 day period. [2]
They were also found not to improve patient outcomes (possibly due to increased number of handoffs, which are highly error prone).
dcreater · 4h ago
Isn't there an obvious conflict of interest when the model maker is also the creator of a benchmark? I think at the very least it should be from a separate business entity under the non profit or from the non profit holding entity itself
gwd · 3h ago
I don't think it's necessarily bad to have the benchmark, but the graphs of Gemini and Claude doing worse than o3 did kind of leave a bad taste in my mouth. "Oh look, your models are worse than ours at this very important metric that we just made up! How terrible!"
progbits · 1h ago
It's also barely better, within the error bar on some metrics. So if this is the best they can do on their own benchmark that's pretty pathetic and shows they have no edge anymore.
iNic · 4h ago
I like that they include the "worst case score at k samples". This is a much more realistic view of what will happen, because someone will get that 1/100 response.
Zaheer · 4h ago
Impressive how well Grok performs in these tests. Grok feels 'underrated' in terms of how much other models (gemini, llama, etc) are in the news.
tough · 4h ago
you can't download grok's weights to run locally
simianwords · 4h ago
how is that relevant here?
tough · 4h ago
it helps explain why theres' less people talking about them than gemini or llama?
less people using them.
reissbaker · 3h ago
You can't download Gemini's weights either, so it's not relevant as a comparison against Gemini.
I think the actually-relevant issue here is that until last month there wasn't API access for Grok 3, so no one could test or benchmark it, and you couldn't integrate it into tools that you might want to use it with. They only allowed Grok 2 in their API, and Grok 2 was a pretty bad model.
tough · 3h ago
lol sorry mixed them up w gemma3 which feels like the open lesser cousin to gemini 2.5/2.0 models
Insanity · 4h ago
I can guarantee you none of my friends (not in tech) use “downloading weights” as an input to select an LLM application.
simianwords · 4h ago
isn't chatgpt the most used or most popular model?
tough · 3h ago
Yes OpenAI has a first-mover advantage and Claude seems to be close as a second player with their closed models too, open weights is not a requirement for success but in an already crowded market (grok's prospect) their preposition isn't competing neither with top tier closed models nor the maybe lesser-capable but more-available battle-tested freely available to run locally open ones
moralestapia · 4h ago
It's not.
Also, only one out of the ten models benchmarked have open weights, so I'm not sure what GP is arguing for.
tough · 3h ago
> in terms of how much other models (gemini, llama, etc) are in the news.
not talking about TFA or benchmarks but the news coverage/user sentiment
...
andy99 · 3h ago
My sense is that these benchmarks are not realistic in terms of the way the model is used. People building specialized AI systems are not, in my experience, letting users just chat with a base model, they would have some variant of RAG plus some guardrails plus other stuff (like routing to pre-written answers for common question).
So what use case does this test setup reflect? Is there a relevant commercial use case here?
dimitri-vs · 5m ago
You could say the same for code benchmarks, no?
For general medical Q&A I can't see how a specialized system would be better than base o3 with web search and a good prompt. If anything RAG and guardrail prompts would degrade performance.
pants2 · 4h ago
This appears to be a very thoughtful and helpful study. It's also impressive to see the improvement in performance in just the last year of model development - almost double.
I've found o3 & deep research to be very effective in guiding my health plan. One interesting anecdote - I got hit in the chest (right over the heart) quite hard a month or so ago. I prompted o3 with my ensuing symptoms and heart rate / oxygenation data from my Apple watch, and it already knew my health history from previous conversations. It gave very good advice and properly diagnosed me with a costochondral sprain. It gave me a timeline to expect (which ended up being 100% accurate) and treatments / ointments to help.
IMO - it's a good idea to have a detailed prompt ready to go with your health history, height/weight, medications and supplements, etc. if anything's happening to you you've got it handy to give to o3 to help in a diagnosis.
quantumwoke · 1h ago
How confident was it in that diagnosis? Did you trust its confidence if so? I asked this to my medico wife and apparently the other possibility is pericardial effusion from injury which can be life threatening.
pizzathyme · 4h ago
Non-clinicians are using ChatGPT every day now to try to find assistance (right or wrong) to real-life medical problems. This is a great evaluation set that could prevent a lot of harm
unsupp0rted · 4h ago
Recently I uploaded a lab report to chatGPT and asked it to summarize it.
It hallucinated serious cancer, along with all the associated details you’d normally find on a lab report. It had an answer to every question I had pre-asked about the report.
The report said the opposite: no cancer detected.
maliker · 3h ago
Interesting. What LLM model? 4o, o3, 3.5? I had horrible performance with earlier models, but o3 has helped me with health stuff (hearing issues).
unsupp0rted · 2h ago
Whichever the default free model is right now- I stopped paying for it when Gemini 2.5 came out in Google's AI lab.
4o, o4? I'm certain it wasn't 3.5
Edit: while logged in
icelancer · 2h ago
> Whichever the default free model is right now
Sigh. This is a point in favor of not allowing free access to ChatGPT at all given that people are getting mad at GPT-4o-mini which is complete garbage for anything remotely complex... and garbage for most other things, too.
Just give 5 free queries of 4o/o3 or whatever and call it good.
pants2 · 2h ago
If you're logged in, 4o, if you're not logged int, 4o-mini. Both don't score well on the benchmark!
askafriend · 2h ago
This gets at the UX issue with AI right now. How's a normie supposed to know and understand this nuance?
unsupp0rted · 2h ago
Or a non-normie. Even while logged in, I had no idea what ChatGPT model it was using, since it doesn't label it. All the label says is "great for everyday tasks".
And as a non-normie, I obviously didn't take its analysis seriously, and compared it to Grok and Gemini 2.5. The latter was the best.
unsupp0rted · 2h ago
Added context: While logged in
maliker · 2h ago
Might be worth trying again with Gemini 2.5. The reasoning models like that one are much better at health questions.
unsupp0rted · 2h ago
Gemini 2.5 in AI Studio gave by far the best analysis
dgfitz · 1h ago
I can’t believe you’re getting downvoted for answering the question about the next-token-predictor model you can’t recall using.
What is happening?
Gracana · 3h ago
I wonder if it was unable to read your report, and just answered as if role-playing?
I gave it a pdf of an engine manual recently and asked some questions, which it answered reasonably. It even pulled a schematic out for me, though it was the wrong one (it gave me a schematic for the CDI ignition variant that we first talked about, rather than the DSAI one we settled on later.)
astura · 2h ago
>I wonder if it was unable to read your report, and just answered as if role-playing?
FWIW,
This is what chatgpt gave me when I uploaded a lab report, it read the results back to me:
Here’s an interpretation of the blood test results you provided:
Other Key Metrics
Absolute Neutrophils, Lymphocytes, Monocytes, Eosinophils, Basophils: All within normal ranges, suggesting no significant abnormalities in immune cell populations.
Anisocytosis (1+): Indicates variation in red blood cell size.
Hypochromasia (1+): Suggests pale red blood cells, supporting iron-deficiency anemia.
Summary
These results suggest iron-deficiency anemia, characterized by low hemoglobin, low MCV, low MCH, and high RDW. The elevated red blood cell count may reflect compensatory mechanisms or mild dehydration.
Your provider recommends a heart-healthy, low-fat diet, which is great for overall health. However, addressing the anemia might also involve increasing dietary iron or taking iron supplements.
The diagnosis is wrong, btw, I don't have iron deficiency. The anemia is caused by a genetic condition called thalassemia, which has been verified by genetic tests. You can use the Mentzer Index to differentiate the two on a simple CBC - https://www.mdcalc.com/calc/10534/mentzer-index
Mine numbers return a "probable diagnosis."
I was wondering if chatgpt would catch it, nope, it didn't. It did say that it was a possibility once I suggested it though.
nradov · 39m ago
It's also disappointing that the LLM would falsely claim that a low-fat diet is "great for overall health". That claim has been thoroughly debunked in the scientific literature (or at least not supported by evidence-based medicine), and yet the recommendation still keeps popping up. While there are certain specific types of fats that can be problematic, in general a moderate amount of fat is fine as a macronutrient.
dekhn · 35m ago
LLMs don't have a magical ability to arbitrate complex truths. If the "low-fat diet" narrative is prominent in the training literature, I would expect the model to say that more often that it said the opposite.
I don't ever really use the term "thoroughly debunked" when referring to nutrition science; as you noted, a better term is that the claim is not supported by the evidence. I've seen enough things debunked and then rebunked to know that nutrition science is not really that accurate.
arcanemachiner · 2h ago
No, cancer detected!
mvdtnz · 2h ago
You're not on Reddit.
arcanemachiner · 1h ago
I typically respect the decorum here, but I couldn't help myself this time.
dekhn · 1h ago
What was the outcome (did the subject of the lab report go on to present cancer)?
(sometimes what we think is a false positive ends up being a true positive later, and cancer is a late-arriving signal).
mrcwinn · 4h ago
Happy to see this. I've struggled with an injury for the past five years. I've been to multiple sports-focused physicians, had various scans. Responses from doctors have ranged from "everything seems fine, can't really figure this out" to [completely wrong hypothesis]. Tried acupuncture. Tried a chiropractor. I remember one doctor, though, had an interesting thought that seemed to make sense - but I've been so discouraged from so many false starts or misplaced hope, I didn't bother following up.
Finally I typed in my entire history into o3-deep-research and let it rip for a while. It came back with a theory for the injury that matched that one doctor, diagrams of muscle groups and even illustrations of proposed exercises. I'm not out of the woods yet, but I am cautiously optimistic for the first time in a long time.
candiddevmike · 3h ago
Why would you trust a LLM over a battery of human experts? I find it hard to believe that the doctors never proposed exercises or some kind of physical therapy for you, at least in the US.
n8henrie · 16m ago
Much of modern medicine's expertise is focused on areas that reimburse well (in some part because those industries subsidize the research).
I would not expect most physicians to have a deep fund of literature-backed knowledge to draw from regarding exercise. Telling someone to do an exercise probably doesn't compensate well.
That said, I'm also pretty negative about the availability of rigorous literature regarding much of nutrition, dentistry, podiatry, physical therapy, etc... you know, the things that affect the health of most human beings that have ever lived.
BeetleB · 2h ago
I can't speak to the OP's condition, but having seen plenty of doctors and physical therapists in the US for over a decade:
Yes, they propose exercises.
No, they don't work.
For certain (common) conditions, PT seems to have it nailed - the exercises really help. For the others, it's just snake oil. Not backed by much research. The current state of the art is just not good when it comes to chronic pain.
So while I don't know if an LLM can be better than a battery of human experts, I do know that those human experts do not perform well. I'm guessing with the OP's case, that battery of human experts does not lead to a consensus - you just end up with 10 different treatments/diagnoses (and occasionally, one is a lot more common than the other, but it's still wrong).
zeroxfe · 2h ago
I've never used LLMs for this, but as someone who's been through a lot of sports-related injuries, I find doctors more or less useless (except for prescribing painkillers and performing surgeries.)
No doctor or physio has ever been able to fix my chronic issues, and I've always had to figure them out myself through lots of self-study and experimentation.
nradov · 33m ago
You have to be realistic and assume that treating most major musculoskeletal injuries will be a lengthy trial-and-error process. Every injury is unique and imaging studies or other tests can only give limited clues about what's going on. There are patients in terrible pain whose scans look normal, and there are other patients whose scans indicate serious pathology yet they are pain-free and completely functional. Instead of getting discouraged, keep trying different modalities (except for the pseudo-science BS) until something works.
In general a lot of those injuries will eventually heal on their own. So it's easy to fool yourself into believing that a particular treatment was effective even when the real cure was time.
quantumwoke · 1h ago
Sounds like you needed a good PT not a physician.
mvdtnz · 2h ago
You trusted known grifters acupuncturists and chiropractors, but not doctors. Then you trusted a computer known for telling lies.
I think you should take a step back and re-assess your internal heuristics.
BigGreenJorts · 1h ago
They literally said they saw sport focused physicians.
scudsworth · 2h ago
i mean it sounds like that doctor had it right then? i dont think this reads as a win for "online self-diagnosis", ai-empowered or otherwise.
lcnPylGDnU4H9OF · 2h ago
From their story, it sounds like the one doctor had it right because the AI-empowered self diagnosis seems to agree. Or do I misunderstand?
Noumenon72 · 3h ago
I hope recent cuts to government science have managed to hit enough of the safetyists and industry captures who keep us from just trying out new healthcare approaches like this and learning. They'd like nothing better than to replace the help you got with "As a large language model, I am unable to offer medical advice."
simianwords · 4h ago
I would really rather like a benchmark purely focusing on diagnosis. Symptoms, patient history vs the real diagnosis. Maybe name this model House M.D 1.0 or something.
The other stuff is good to have but ultimately a model that focuses on diagnosing medical conditions is going to be the most useful. Look - we aren't going to replace doctors anytime soon but it is good to have a second opinion from an LLM purely for diagnosis. I would hope it captures patterns that weren't observed before. This is exactly the sort of thing game that AI can beat a human at - large scale pattern recognition.
srameshc · 4h ago
Is the Med-PaLM model that Google's has been working on meant to be considered for comparison ? If I'm not mistaken, it isn't publicly available.
Med-PaLM is old and has been superseded by (multiple generations of) Gemini.
NKosmatos · 1h ago
Most probably I’m going to get downvoted, but I’m gonna say it…
It’s a pity they don’t support Greek language, keeping in mind that almost all medical terminology has Greek origins.
Anyhow, this is a step in the good direction and for sure it will aid many people looking for medical assistance via ChatGPT.
GuinansEyebrows · 4h ago
i have zero trust in openai's ability to do anything impartially. why should we leave the judgement of a private tool up to the makers of the tool especially when human lives are at stake?
beezlebroxxxxxx · 4h ago
I can already see the pharma salesmen drooling at the idea of how various symptoms can be marketed to.
"As an AI agent, a possible diagnosis is [xxx]. Ask your doctor about / look into [yyy™] for a possible solution!"
ceejayoz · 3h ago
And OpenAI is definitely thinking about this on their end:
> OpenAI CFO Sarah Friar told the Financial Times that OpenAI is weighing the inclusion of ads in its products but wants to be "thoughtful about when and where we implement them."
barnas2 · 3h ago
Ad spots inside chatgpt are going to be worth an obscene amount of money.
simianwords · 4h ago
I agree - we should exercise a bit of caution here. There is no way they would release a benchmark which makes their model look bad. But then again we know that their models are one of the best for other uses so its not a big leap to accept this benchmark.
amarcheschi · 3h ago
I think that the damage of "chatgpt misdiagnose X as Y, person dies of Z" would be quite bad for PR
yapyap · 1h ago
Sam Altman does not care about “improving human health”
kypro · 3h ago
Why are all the label colours for the "Worst-case HealthBench score at k samples" chart the same colour and the same shape? Completely unreadable.
brap · 1h ago
The colors are slightly different. I think the person why made this was more concerned with appearance and not providing information.
ramon156 · 4h ago
I don't want to be a conspiracy theorist, but could this be in preparation for Amazon's (to be) health branch?
With the healthcare prices increasing at the breakneck speed, I am sure AI will take more and more role in diagnosing and treating people's common illnesses, and hopefully (doubt it), the some of that savings will be transferred to the patients.
P.S. In contrast to the US system, in my home city (Rangoon, Burma/Myanmar), I have multiple clinics near my home and a couple of pharmacy within two bus stops distance. I can either go buy most of the medications I need from the pharmacy (without prescription) and take them on my own (why am I not allowed to take that risk?) OR I can go see a doctor at one of these clinics to confirm my diagnosis, pay him/her $10-$20 for the visit, and then head down to the pharmacy to buy the medication. Of course, some of the medications that include opioids will only be sold to me with the doctor's prescription, but a good number of other meds are available as long as I can afford them.
The trouble is you are not educated enough to tell what is simple and what isn't. A cough could be a cough or it could be something more serious, only a "real" examination will reveal that. And sometimes even that's not enough, you need an examination by a specialists.
I'll tell you a story. Once upon a time I got pain in my balls. I went to a doctor and he felt around and he said he didn't feel anything. I went to another doctor and he felt something, but he had no idea what it was. He said could be a cyst, could be a swollen vein, could be an infection - he didn't even know if it was on the testicle or on the tube thingy.
Then I went to a Urologist. You can tell this man has felt up a lot of balls. He felt me up and said, "yup, that's a tumor" almost immediately. He was right, of course, and he ended up being the one to remove it too. Since I caught the cancer pretty early the chemotherapy wasn't too intense.
Point is, expertise matters when things aren't straight forward. Then, experience and perspective gets to shine.
Because you're paying for the expertise of someone who studied for more than a decade which you won't get from a random web search.
An AI system with today's technology should be less trustworthy for medical diagnosis than a web search. At least with a web search you might stumble upon a site with content from experts, assuming you trust yourself to be able to discern expert advice from bot-generated and spam content. Even if a doctor is doing the searching instead of me, I would pay them only for their knowledge to make that discernment for me. Why you think an AI could do better than a human at that is beyond me.
Your question reminds me of that famous Henry Ford GE invoice story:
> Making chalk mark on generator: $1.
> Knowing where to make mark: $9,999.
> Why you think an AI could do better than a human at that is beyond me.
Why do you think an AI couldn't do better than a human, when we have ample evidence of computers/AI exceeding humans in many areas?
We have a massive, massive shortage of doctors.
The industry is doing everything they can to make it worse by the day, so I won't hold my breath that we'll get the slightest bit of respite.
It'd obviously be ideal if everyone could see a doctor for an affordable price any time they wanted.
We don't live in the ideal world.
This would be a HUGE win for most people.
https://www.bls.gov/ooh/healthcare/physician-assistants.htm
This insanity needs to be regulated yesterday.
Because while LLMs obviously have massive limitations, so do humans, and it's not entirely clear to me that some synthesis of the two can't produce much better results than either on its own.
In theory, I agree with you. The world "some" is doing a lot of heavy lifting there though. I only hope that whatever definition of some emerges, it's not a horribly flawed one.
[1] https://psnet.ahrq.gov/primer/duty-hours-and-patient-safety
[2] https://www.fmcsa.dot.gov/sites/fmcsa.dot.gov/files/docs/Dri...
They were also found not to improve patient outcomes (possibly due to increased number of handoffs, which are highly error prone).
less people using them.
I think the actually-relevant issue here is that until last month there wasn't API access for Grok 3, so no one could test or benchmark it, and you couldn't integrate it into tools that you might want to use it with. They only allowed Grok 2 in their API, and Grok 2 was a pretty bad model.
Also, only one out of the ten models benchmarked have open weights, so I'm not sure what GP is arguing for.
not talking about TFA or benchmarks but the news coverage/user sentiment ...
So what use case does this test setup reflect? Is there a relevant commercial use case here?
For general medical Q&A I can't see how a specialized system would be better than base o3 with web search and a good prompt. If anything RAG and guardrail prompts would degrade performance.
I've found o3 & deep research to be very effective in guiding my health plan. One interesting anecdote - I got hit in the chest (right over the heart) quite hard a month or so ago. I prompted o3 with my ensuing symptoms and heart rate / oxygenation data from my Apple watch, and it already knew my health history from previous conversations. It gave very good advice and properly diagnosed me with a costochondral sprain. It gave me a timeline to expect (which ended up being 100% accurate) and treatments / ointments to help.
IMO - it's a good idea to have a detailed prompt ready to go with your health history, height/weight, medications and supplements, etc. if anything's happening to you you've got it handy to give to o3 to help in a diagnosis.
It hallucinated serious cancer, along with all the associated details you’d normally find on a lab report. It had an answer to every question I had pre-asked about the report.
The report said the opposite: no cancer detected.
4o, o4? I'm certain it wasn't 3.5
Edit: while logged in
Sigh. This is a point in favor of not allowing free access to ChatGPT at all given that people are getting mad at GPT-4o-mini which is complete garbage for anything remotely complex... and garbage for most other things, too.
Just give 5 free queries of 4o/o3 or whatever and call it good.
And as a non-normie, I obviously didn't take its analysis seriously, and compared it to Grok and Gemini 2.5. The latter was the best.
What is happening?
I gave it a pdf of an engine manual recently and asked some questions, which it answered reasonably. It even pulled a schematic out for me, though it was the wrong one (it gave me a schematic for the CDI ignition variant that we first talked about, rather than the DSAI one we settled on later.)
FWIW,
This is what chatgpt gave me when I uploaded a lab report, it read the results back to me:
Here’s an interpretation of the blood test results you provided:
White Blood Cell Count (WBC) Value: 9.2 (Normal: 3.8 - 10.8 Thousand/uL)
Within the normal range, indicating no immediate signs of infection or inflammation.
Red Blood Cell Count (RBC) Value: 5.29 (Normal: 3.8 - 5.1 Million/uL)
Elevated (High). This could indicate dehydration, stress, or other conditions like polycythemia.
Hemoglobin Value: 10.9 (Normal: 11.7 - 15.5 g/dL)
Low. Suggests anemia, possibly due to iron deficiency or chronic disease.
Hematocrit Value: 36.0% (Normal: 35.0 - 45.0%)
Within the normal range, but on the lower side, supporting the presence of mild anemia.
Mean Corpuscular Volume (MCV) Value: 68.1 (Normal: 80.0 - 100.0 fL)
Low. Indicates microcytic anemia, often due to iron deficiency.
Mean Corpuscular Hemoglobin (MCH) Value: 20.6 (Normal: 27.0 - 33.0 pg)
Low. Suggests hypochromia (paler red blood cells), often associated with iron-deficiency anemia.
Mean Corpuscular Hemoglobin Concentration (MCHC) Value: 30.3 (Normal: 32.0 - 36.0 g/dL)
Slightly low. May reflect reduced hemoglobin in red blood cells, consistent with iron-deficiency anemia.
Red Cell Distribution Width (RDW) Value: 15.8% (Normal: 11.0 - 15.0%)
Elevated. Indicates variability in red blood cell size, common in iron-deficiency anemia.
Platelet Count Value: 397 (Normal: 140 - 400 Thousand/uL)
Normal, but on the higher side of the range.
Other Key Metrics Absolute Neutrophils, Lymphocytes, Monocytes, Eosinophils, Basophils: All within normal ranges, suggesting no significant abnormalities in immune cell populations.
Anisocytosis (1+): Indicates variation in red blood cell size.
Hypochromasia (1+): Suggests pale red blood cells, supporting iron-deficiency anemia.
Summary These results suggest iron-deficiency anemia, characterized by low hemoglobin, low MCV, low MCH, and high RDW. The elevated red blood cell count may reflect compensatory mechanisms or mild dehydration.
Your provider recommends a heart-healthy, low-fat diet, which is great for overall health. However, addressing the anemia might also involve increasing dietary iron or taking iron supplements.
The diagnosis is wrong, btw, I don't have iron deficiency. The anemia is caused by a genetic condition called thalassemia, which has been verified by genetic tests. You can use the Mentzer Index to differentiate the two on a simple CBC - https://www.mdcalc.com/calc/10534/mentzer-index
Mine numbers return a "probable diagnosis."
I was wondering if chatgpt would catch it, nope, it didn't. It did say that it was a possibility once I suggested it though.
I don't ever really use the term "thoroughly debunked" when referring to nutrition science; as you noted, a better term is that the claim is not supported by the evidence. I've seen enough things debunked and then rebunked to know that nutrition science is not really that accurate.
Finally I typed in my entire history into o3-deep-research and let it rip for a while. It came back with a theory for the injury that matched that one doctor, diagrams of muscle groups and even illustrations of proposed exercises. I'm not out of the woods yet, but I am cautiously optimistic for the first time in a long time.
I would not expect most physicians to have a deep fund of literature-backed knowledge to draw from regarding exercise. Telling someone to do an exercise probably doesn't compensate well.
That said, I'm also pretty negative about the availability of rigorous literature regarding much of nutrition, dentistry, podiatry, physical therapy, etc... you know, the things that affect the health of most human beings that have ever lived.
Yes, they propose exercises.
No, they don't work.
For certain (common) conditions, PT seems to have it nailed - the exercises really help. For the others, it's just snake oil. Not backed by much research. The current state of the art is just not good when it comes to chronic pain.
So while I don't know if an LLM can be better than a battery of human experts, I do know that those human experts do not perform well. I'm guessing with the OP's case, that battery of human experts does not lead to a consensus - you just end up with 10 different treatments/diagnoses (and occasionally, one is a lot more common than the other, but it's still wrong).
No doctor or physio has ever been able to fix my chronic issues, and I've always had to figure them out myself through lots of self-study and experimentation.
In general a lot of those injuries will eventually heal on their own. So it's easy to fool yourself into believing that a particular treatment was effective even when the real cure was time.
I think you should take a step back and re-assess your internal heuristics.
The other stuff is good to have but ultimately a model that focuses on diagnosing medical conditions is going to be the most useful. Look - we aren't going to replace doctors anytime soon but it is good to have a second opinion from an LLM purely for diagnosis. I would hope it captures patterns that weren't observed before. This is exactly the sort of thing game that AI can beat a human at - large scale pattern recognition.
> https://sites.research.google/med-palm/
It’s a pity they don’t support Greek language, keeping in mind that almost all medical terminology has Greek origins.
Anyhow, this is a step in the good direction and for sure it will aid many people looking for medical assistance via ChatGPT.
"As an AI agent, a possible diagnosis is [xxx]. Ask your doctor about / look into [yyy™] for a possible solution!"
https://www.axios.com/2024/12/03/openai-ads-chatgpt
> OpenAI CFO Sarah Friar told the Financial Times that OpenAI is weighing the inclusion of ads in its products but wants to be "thoughtful about when and where we implement them."