Study: ChatGPT Outperforms Physicians in Quality, Empathetic Answers to Patients

Re:Study: ChatGPT Outperforms Physicians in Quality, Empathe... by jasondigitized on Saturday April 29, 2023 @10:02AM (#35752941)

This is the exact opposite experience I have had with code. While it isn’t perfect it’s ability to scaffold code is a huge productivity booster. Example. Tailwind. “How can I highlight a parent container when I hover on any of its children using Tailwind”. JavaScript “How can I merge two JavaScript objects and remove any duplicates”.

[ Parent | on HN ]

Re:Study: ChatGPT Outperforms Physicians in Quality, Empathe... by lm28469 on Saturday April 29, 2023 @10:26AM (#35753162)

Yeah it's very good at answering basic questions which have been answered 50+ times on stack overflow &co that's for sure.

I use it for my side projects, for tech I have no experience in, and it works very well, because I know what I want, I know that it is possible and I just need it to vomit the boilerplate to save me 5 google searches

For my day job it's next to useless, and if your day job can already be automated by chatgpt I have bad news for you

[ Parent | on HN ]

Re:Study: ChatGPT Outperforms Physicians in Quality, Empathe... by 2devnull on Saturday April 29, 2023 @12:07PM (#35754215)

“if your day job can already be automated by chatgpt”

If they can automate the work of a physician, who exactly is safe? Low skill labor, maybe, for awhile.

[ Parent | on HN ]

Re:Study: ChatGPT Outperforms Physicians in Quality, Empathe... by meroes on Saturday April 29, 2023 @12:33PM (#35754453)

The remote physician misdiagnosed my X-ray last time. That physician is easily automated out and possibly for the benefit of the patients not just costs. The other staff involved, like the NP, X-ray tech, assistant, are fine for a lot longer.

[ Parent | on HN ]

Re:Study: ChatGPT Outperforms Physicians in Quality, Empathe... by 2devnull on Sunday April 30, 2023 @10:10AM (#35762700)

“are fine for a lot longer.”

My point is that “a longer” implies “when” rather than “if”. Everyone is replaceable, in the long run.

[ Parent | on HN ]

Re:Study: ChatGPT Outperforms Physicians in Quality, Empathe... by lm28469 on Saturday April 29, 2023 @03:31PM (#35756277)

> If they can automate the work of a physician, who exactly is safe?

But they can't, it's like saying you're 15cl stove top italian coffe maker is replacing a starbucks tier coffee machine

If the only metric you account for is "it makes coffee" boolean then sure, if you actually implement it you'll notice things falling apart in the first 10 minutes.

[ Parent | on HN ]

Re:Study: ChatGPT Outperforms Physicians in Quality, Empathe... by logifail on Saturday April 29, 2023 @10:45AM (#35753344)

> JavaScript “How can I merge two JavaScript objects and remove any duplicates”

(This is a genuine question) Is it hard to find a solution to that question on the web by using a search engine?

[ Parent | on HN ]

Re:Study: ChatGPT Outperforms Physicians in Quality, Empathe... by ericmcer on Saturday April 29, 2023 @12:07PM (#35754217)

It would be difficult to search because that is a weird question. If you asked a more practical question like "merge two arrays and remove duplicate values" you would find tons of exact results.

[ Parent | on HN ]

Re:Study: ChatGPT Outperforms Physicians in Quality, Empathe... by catlifeonmars on Saturday April 29, 2023 @12:41PM (#35754534)

* * *

[ Parent | on HN ]