Operation Spiderweb: https://en.wikipedia.org/wiki/Operation_Spiderweb |
PP here. I'm not completely against the rise in AI diagnoses. Doctors do a terrible job of diagnosing women and minorities, for instance. The average of 7-11 years to diagnose endometriosis is a disgrace and comes down primarily to doctors' disrespect for women's pain and disdain for women's self-reported symptoms. But the improvements in patient care will come at the expense of doctors' bottomline. 7-11 years of an endometriosis patient going to doctor visit after doctor visit and receiving useless procedures, diagnoses, and prescriptions is a lot of money in doctors' pockets. If all of that could be wrapped up in a couple of visits to take a detailed patient history followed by plugging things into AI, that's a lot of money lost. Imagine that multiplied across thousands of health conditions for millions of patients annually. |
Yeah, I think the guy making Excell spreadsheets is the one that can actually use AI effectively. That's the problem. AI is more like a multiplier effect. Those that were already good at data science have just become that much more effective; they can literally automate all of the corporate ladder climbers. |
No. I appreciate you have family members who are doctors, but I am a doctor. AI isn't encroaching significantly onto healthcare. Sorry. Cite it, if you think you have other reasons besides anecdote. OpenEvidence is helpful. There are places in radiology and pathology reporting where a first pass with AI is sometimes helpful. Where else? That being said, I'm a doctor. Of course, if there is a better way to provide health care to people, I'm all for it. I'm sure we will get there eventually. Heck, if there were no illnesses, accidents, or aging, I'd be out of a job -- and that would be great. Putting me out of work would be a fantastic endgame. Until then, I'm not going to lose my mind, soil myself, or whatever it is you think I should be doing at the prospect. Not going to change the outcome anyway. |
There's an insane shortage of providers right now. More efficiency, if it actually helped and was reliable, would alleviate that somewhat. The current docs wouldn't suffer from lack of patients. We might need fewer doctors in the longterm, but people applying for medical school would know it by then. This isn't an argument, and it isn't a problem. |
So, you say we're subsidizing healthcare for the rich, and now they won't be able to afford their expensive procedure, maybe there is a god. Unfortunately, I doubt this kind of functionality is going to be in AI for very long, they'll figure out how to kill it like the free and open internet, so enjoy it while it lasts. |
Offshoring had a similar bumpy start, with all sorts of quality control issues. Some industrial work has also remained domestically, but only a fraction of what used to be. But in 2-3 decades a ton of white-collar work will get replaced by automation of one sort or another. We should be celebrating this, because who wants to live the Accounts Receivable life? However our capitalist betters will make sure this transition is just as miserable for us as the off-shoring and migrant labor transitions were. |
Healthcare is for profit in the US. How are these unemployed people going to afford AI healthcare? |
Ask the Saskatchewan wheat farmers of the 1940s. |
You must own AI stock, because there’s no way you can say this with any kind of authority. |
What kind of doctor are you and how long ago did you complete your residency? Your response is completely out of touch and you're clearly not staying up on recent developments in medicine. |
I keep advocating for the AMA and state licensing authorities to make continuing medical education requirements at least as rigorous as continuing legal education requirements. The way doctors who aren't in hospital and research settings quickly lose track of new drugs, interventions, and advances is disturbing. Anyway, let me catch you up: https://www.nytimes.com/2024/11/17/health/chatgpt-ai-doctors-diagnosis.html https://www.wired.com/story/microsoft-medical-superintelligence-diagnosis/ https://time.com/7299314/microsoft-ai-better-than-doctors-diagnosis/ https://hms.harvard.edu/news/ai-system-detailed-diagnostic-reasoning-makes-its-case https://www.newyorker.com/magazine/2025/09/29/if-ai-can-diagnose-patients-what-are-doctors-for https://newsnetwork.mayoclinic.org/discussion/mayo-clinics-ai-tool-identifies-9-dementia-types-including-alzheimers-with-one-scan/ |
Okay, I'll answer your questions after you answer the question I already asked: where else (other than pathology and radiology, and that as just first pass) is AI "encroaching significantly onto healthcare?" Again -- cite it, if you think you have other reasons for believing this besides "my dad said" anecdotes. |
You count small individual studies as "encroaching significantly onto healthcare?" Okay, there's your problem. Small studies are just studies. Nobody is paying for any of this instead of actual people providing healthcare. Nothing has been replaced.
I'm a pediatrician who has spent 25 years in this business, working across two countries and teaching in three medical schools. Of course I know about the studies you referenced, but you haven't made a case for them significantly "encroaching" onto anything except clickbait real estate. Try again. |
PS: You should take a course in how research works. Seriously. |