Anyone else dooming about AI?

physician

You mean being a doctor or doing physics
hi sam altman

I mean a doctor

Only nurses, EMT and surgeons are safe. My recommendation is you look into becoming an EMT if you want to have job security and actually help people.

thats lefty nonsense. they think healthcare is the only industry that matters for some reason. probably because they all have aids.

EMT pay is God awful though

nurses

safe

You're retarded also you're forgetting all the people whose job it is to know how to use certain medical equipment for example, ultrasound technician.

Someone has to lift the old people anon and all the specialized technicians will be standardized to the point nurses will be able to handle almost everything that doesnt involve lab work

Pay or fulfillment. If the family is well off then it shouldnt matter.

My job is not gonna be replaced by Ai but nonetheless the competition is too high for me to get hired kek

You mean being a doctor or doing physics

Doing physics makes you a physicist, not a physician

Nurses can't even handle giving someone an IV lol

moving old people

Weird you think AI will take over the medical industry but there will still be a need for "nurses" to move a person from one bed to another.

Who whole medical field won't be replaced ill people want to talk to real people and will gladly pay for it
as an esl yuropoor i sometimes mix up things

I sometimes think the whole AGI craze is done by already employed tech guys knowing it won't ever come but wanting to thin out their competition

They want you scared of AI. It's intentional. Whether or not it works is inconsequential.

The more job seekers that are scared the easier it is to hire them and the less they have to be paid.

They fear a worker's market where employees can start demanding higher salaries and will spend billions to avoid it.

rfk jr just went in front of Congress a week ago and said he wants nurses replaced with ai right now kek.

People who have no idea what a physician does will come up with shit like this with such confidence, it's hilarious. Is ChatGPT going to do chest or ascitic drains, endoscopies, suture, fit central or peripheral lines? Intubate? Run a cath lab? No.

Is it going to take legal responsibility for the diagnosis and management of a patient? No.

Is going to manage acute presentations? Do you think some dipshit using it is going to do that either?

crazy, I hadn't seen that

s ChatGPT going to do chest or ascitic drains, endoscopies, suture, fit central or peripheral lines? Intubate? Run a cath lab? No.

This is the main thing keeping me hopeful. Robotics still have a long way to go before we're even close to that.

the bottle neck for blue collar work (and also surgeons specifically) is finger dexterity. theres no real breakthroughs on the horizon here but whoever solves it will officially end the old world. once this happens all physical tasks will be done by a robot. the cost of things like hair transplants and breast surgery drops down to basically nothing. healthcare gets extremely cheap.
there really isn't a bottle neck for white collar work. that is more or less here already.

its for the best. the cheaper and cheaper we can make healthcare the easier it will be to finally get free healthcare.

Robotics and machine learning won't save you. Also even if it somehow got cheaper you would still pay the same amount lol

cheaper healthcare costs basically makes it easier for the government to foot the bill. basically there would be no argument that we shouldn't have free healthcare.

What a nice image...if it wasn't ai it would be something else. Maybe it will finally create socialism in the US.

procedures are the main thing keeping me hopeful

Why? If you want to go to medical school, maybe find out what a doctor does. You can't just give a retard a stethoscope + ChatGPT and expect a timely diagnosis; even in primary care this just won't work.

Think about it. I can type in "2-year-old presents with barking cough and stridor" verbatim and ChatGPT will tell me it's croup, and how to manage croup. I already know that. Why? Because I know how to recognize and describe the symptoms of croup; because I'm trained to recognize and manage this presentation. What does the LLM add to this process? Fuck all, and if you couldn't recognize stridor, you wouldn't get the diagnosis out of it. If I said "55-year-old male presents with tearing chest pain and hemodynamic instability", the patient will die while I read through ChatGPT's management plan, which would be accurate but again useless, I already know it and I'm on the phone to surgery while a reg is infusing labetalol. You need someone who knows what to do immediately and who is going to take legal responsibility for those decisions. LLMs add nothing to the process; the most they're going to do is sift through imaging, or help patients better describe their presentations.

The only field I can think of that might be wiped out by AI in our lifetime is pathology, and even then, they have academia to fall back on.

a nurse could do everything you just described. no need to be a full doctor to recognize someone is having a heart attack. and a nurse with chat gpt can completely cut you out of the picture making healthcare cheaper and more accessible for everyone.

If this was the case, then hospitals would already just fire all the doctors because they're expensive and just hire nurses

heart attack

I was describing aortic dissection you absolute dipshit, and even if it were an MI, who treats that?

what i said to you frightened you

they will definitely be downsized more

nurses will run the cath labs... with ChatGPT!

:^)

a nurse could do everything you just described

Uh no. A nurse that went to medical be a doctor but was too cool for exams. The average 90 IQ brown nurse? lmao

after reading what that it is thats exactly the kind of scutwork that will not require a full on doctor in the future. a little more than a nurse with chat gpt actually can do a lot more than you are willing to admit because it terrifies you

Paper by physicians at Harvard and Stanford: "In all experiments, the LLM displayed superhuman diagnostic and reasoning abilities."

reddit.com/r/artificial/comments/1kyi09b/paper_by_physicians_at_harvard_and_stanford_in/

you're fucked

Why aren't they downsized right now?

As an autist, I was already a useless NEET before the 2008 cataclysm made it impossible to get a sustainable job.
At this point, I hope it happens fast, so they are fucking forced to enact a UBI and Keynes 15 hour work week.
Too much societal damage will be done if it does not happen. Dumb elitist bastards decided to break capitalism with 'globalism for the lowest bidder', so they will have to fucking pony up from all the billions they fucking stole in low wages and profiteering these past 25-55 years. CUNTS!

don't worry they will be. and thats ultimately better for society even if its bad for you can't you understand that

a nurse is going to diagnose the location of an infarct, sedate the patient, Seldinger a catheter into the femoral artery, through the external iliac, all the way up the aorta, into the correct coronary artery, and clear a thrombus, using real-time radiography, while stabilizing them throughout the procedure, and then medically manage the patient until discharge

+/- fitting and calibrating a pacemaker

... because ChatGPT!

If you asked a nurse if they wanted to do this, they'd tell you to fuck yourself.

don't worry they will be.

Explain to me why they aren't downsized right now and the best thing you can give me is muh two weeks

ultimately better for society

How is it better when in the end you'll pay exact the same amount while whoever owns the hospitals is happy that they have to pay less expenses
Im what form was rhe data provided to the llms? Im too lazy to read the whole paper

yeah but a full on doctor is not required here. something maybe more than a nurse but less than a doctor can do this, up until robotics solves finger dexterity and then your whole field is 100% gone

at least read the highlighted text here you dickhead

Im asking about the actual data not the abstract

The highlighted is their results not their abstract, actual data you'll have to read the whole paper

The highlighted is their results not their abstract

do you understand what an abstract is?

actual data you'll have to read the whole paper

so you didn't look at the actual data yourself and just screamed the proclaimed results without looking at the data, methodology and results?

you're claiming that harvard and stanford fudged their results? thats your cope? these people said they want protected trials to begin immediately

you're claiming that harvard and stanford fudged their results?

considering that the paper is just on arXiv, yes. im just questioning if they handed in the raw conversation with the patient or did something with the data (like extracting all the symptoms) before handing it in to the llm

yeah but a full on doctor is not required here

What would your procedure monkey do if a free-wall rupture occurred during the procedure?
The LLMs were presented clinical vignettes, which is the same shit I mentioned earlier, i.e., you type in

2-year-old presents with rapid-onset fever, drooling, peripheral cyanosis, sats of 90%

I just typed that into ChatGPT, and it gets acute epiglottitis, which is correct. The management plan is "give humidified oxygen and call Daddy (the on-call anesthetist)". The nurses already know this, it's already done.

So how do we further streamline this process, without throwing patient safety out the window? Do we really want Nurse Jackie to have a go at rapidly intubating a deteriorating 2-year-old?

you're claiming that harvard and stanford fudged their results?

Yes, especially when it comes to AI. They're university students and cloistered academics, not senior developers with real-world experience.
Idiots like this keep trying to use AI in court with disastrous results. If AI can pass the bar and not be a good lawyer, it can look like a doctor without being a good doctor. Five bucks says that I could get it to make up a body part.

Simply reading the results of a study and nothing else is really, really dumb.

did you read any of the paper before leaping to your biased conclusion

2-year-old presents with rapid-onset fever, drooling, peripheral cyanosis, sats of 90%

Id assume the actual hard part is to get all those symptoms out of some patient

There exists other posters on this board besides you and the ones you argue with.

Like how everyone lost their jobs when computers took over?
stop being a faggot you are just looking for an excuse to not even try when even when the resources you have available to you make getting a degree easier then ever. When I went to college I didn't have tools which could write essays for me, Stop being a bitch.

extreme situations such as something requiring surgery requires a surgeon. but your black and white examples are what you are latching onto because you don't want to admit all the other contexts where you will not be needed and therefore less money needing to end up going into your pockets - your greatest fear. think of all the situations where you are redundant instead of the few extremes that you aren't

once it gets that data its reasoning skills are beyond human
and these med students in here acting like that doesn't make their asshole quiver

noooo what about my precious moneyyyyy

once it gets that data

And how do you get the data out of the patient? Imagine you have either some borderline retarded guy, some senile man or a screaming kid. The individual parts mean nothing if you can't integrate them into the system

Anyone can wear smart-glasses integrated with chatgpt which understands and can describe what is going on in the camera frame and then provide instructions. Now anyone can become a doctor and carry out the physical actions required. Then when robots get good enough, humans are out of the picture entirely

i would imagine a tiered system based on severity. a nurse with ai looks at your sprained with no doctor present ever and doctor (also probably using the same ai) sees you when your feeling dizzy and don't know why
however thats a more reduced role, something more than a nurse but less than a full doctor

extreme situations such as something requiring surgery requires a surgeon

Yet MI is managed by interventional cardiologists, not surgeons.

think of all the situations where you are redundant instead of the few extremes that you aren't

These are extreme scenarios to you.

Anyone can wear smart-glasses integrated with chatgpt which understands and can describe what is going on in the camera frame and then provide instructions

Again if this is allegedly possible why does nobody use it then?
Adding more overhead to an already complex system is silly

i would imagine a tiered system based on severity.

Bro is literally describing triage.

It is pretty worrying how school now is just people doing AI generated assignments with AI.

First world countries need more simple workers and that's a good way to ensure that there are enough of them

Yeah, I am so glad everyone in the US will go back to working in sweatshops. I cannot wait to work in the iPhone factory. Thank you chatgpt.