Another ChatGPT answer in QA (Sean PT)
-
EDIT by Nelek: Nothing to see here (but Sean's answer below), move on, move on... Sean PT - https://www.codeproject.com/script/Membership/View.aspx?mid=15600406 Member questions & answers However, this one: a) openly states that it was generated by ChatGPT; and b) is an answer to a no-code homework question. So I'm not sure whether that deserves a kicking for the user. Any thoughts?
"These people looked deep within my soul and assigned me a number based on the order in which I joined." - Homer
"These people looked deep within my soul and assigned me a number based on the order in which I joined" - Homer
-
EDIT by Nelek: Nothing to see here (but Sean's answer below), move on, move on... Sean PT - https://www.codeproject.com/script/Membership/View.aspx?mid=15600406 Member questions & answers However, this one: a) openly states that it was generated by ChatGPT; and b) is an answer to a no-code homework question. So I'm not sure whether that deserves a kicking for the user. Any thoughts?
"These people looked deep within my soul and assigned me a number based on the order in which I joined." - Homer
Richard Deeming wrote:
So I'm not sure whether that deserves a kicking for the user. Any thoughts?
As the source of the solution is given, I don't see it as a problem. It is no different from pointing to StackOverflow or Blog posts. I have an issue when ChatGPT answers are claimed as their own work. This is plagiarism, no different from claiming someone else's answer from StackOverflow or a blog post as their own. However, helping someone with their homework is a concern. If they use it for their homework and learn nothing, they're only hurting themselves. What I try to do, most of the time is teaching members how to research their answers, and give them the skill to help them learn and grow.
Graeme
"I fear not the man who has practiced ten thousand kicks one time, but I fear the man that has practiced one kick ten thousand times!" - Bruce Lee
-
EDIT by Nelek: Nothing to see here (but Sean's answer below), move on, move on... Sean PT - https://www.codeproject.com/script/Membership/View.aspx?mid=15600406 Member questions & answers However, this one: a) openly states that it was generated by ChatGPT; and b) is an answer to a no-code homework question. So I'm not sure whether that deserves a kicking for the user. Any thoughts?
"These people looked deep within my soul and assigned me a number based on the order in which I joined." - Homer
Please leave this guy alone. He is an integrator for CodeProject.AI Server.
Thanks, Sean Ewington CodeProject
-
Please leave this guy alone. He is an integrator for CodeProject.AI Server.
Thanks, Sean Ewington CodeProject
Maybe there should be some way of indicating that sort of thing in the profile? But obviously something that can't be faked by the user. :) Also, is there an official policy yet on using GPT to generate answers?
"These people looked deep within my soul and assigned me a number based on the order in which I joined." - Homer
-
Maybe there should be some way of indicating that sort of thing in the profile? But obviously something that can't be faked by the user. :) Also, is there an official policy yet on using GPT to generate answers?
"These people looked deep within my soul and assigned me a number based on the order in which I joined." - Homer
Funny you should say that because we are working on making them more identifiable. I'm still trying to figure out how we could conclusively determine someone was using GPT to generate answers.
Thanks, Sean Ewington CodeProject
-
Funny you should say that because we are working on making them more identifiable. I'm still trying to figure out how we could conclusively determine someone was using GPT to generate answers.
Thanks, Sean Ewington CodeProject
Sean Ewington wrote:
I'm still trying to figure out how we could conclusively determine someone was using GPT to generate answers.
Assuming this tool[^] wouldn't work? :) Obviously at the moment it's a judgement call. But doesn't it makes sense to have a formal policy in place that we can point to? Otherwise, we risk either deleting answers which are actually acceptable, or getting complaints from users who "didn't realise" that their behaviour was against the policy, because there is no policy.
"These people looked deep within my soul and assigned me a number based on the order in which I joined." - Homer
-
Sean Ewington wrote:
I'm still trying to figure out how we could conclusively determine someone was using GPT to generate answers.
Assuming this tool[^] wouldn't work? :) Obviously at the moment it's a judgement call. But doesn't it makes sense to have a formal policy in place that we can point to? Otherwise, we risk either deleting answers which are actually acceptable, or getting complaints from users who "didn't realise" that their behaviour was against the policy, because there is no policy.
"These people looked deep within my soul and assigned me a number based on the order in which I joined." - Homer
That tool requires a minimum of 1000 characters. So sometimes it might work, sometimes it might not. I'll start using it and see how it goes. How about for now, until we can definitely prove someone is using ChatGPT, their accounts can stay and their answers can simply get downvoted to oblivion.
Thanks, Sean Ewington CodeProject
-
That tool requires a minimum of 1000 characters. So sometimes it might work, sometimes it might not. I'll start using it and see how it goes. How about for now, until we can definitely prove someone is using ChatGPT, their accounts can stay and their answers can simply get downvoted to oblivion.
Thanks, Sean Ewington CodeProject
Sean Ewington wrote:
until we can definitely prove someone is using ChatGPT, their accounts can stay
Wouldn't that actively discourage users from admitting to the source of their answer(s) though? :~ Eg: Someone posting a 500-word GPT-generated answer every 30 seconds without attribution gets to stay because we can't prove how the answers are being generated, whereas someone posting two or three a day with attribution gets kicked. That seems like it's the wrong way round to me.
"These people looked deep within my soul and assigned me a number based on the order in which I joined." - Homer
-
Sean Ewington wrote:
until we can definitely prove someone is using ChatGPT, their accounts can stay
Wouldn't that actively discourage users from admitting to the source of their answer(s) though? :~ Eg: Someone posting a 500-word GPT-generated answer every 30 seconds without attribution gets to stay because we can't prove how the answers are being generated, whereas someone posting two or three a day with attribution gets kicked. That seems like it's the wrong way round to me.
"These people looked deep within my soul and assigned me a number based on the order in which I joined." - Homer
Sorry for the confusion. I'm not suggesting we kick people who are admitting they are using it. Those people can communicated with. If communication fails, and they're an endless terror, then we kick.
Thanks, Sean Ewington CodeProject
