Earlier quoted context omitted.
So a Hacker News comment.
Yep, they trained the AI on hackernews
Oh no, not at all. In fact, hackernews got trained on the AI output of the Previous Turn.
Use of ChatGPT generated text for posts on Stack Overflow is temporarily banned
81–90 of 143 posts
Re: Use of ChatGPT generated text for posts on Stack Overflow is temporarily banned
#82A solution should have a parallel answers thread, system-gen vs bio-gen
Re: Use of ChatGPT generated text for posts on Stack Overflow is temporarily banned
#83What is the mechanism by which a text is determined to originate from ChatGPT? This will be crucial for preventing unfavorable training loops among a myriad of other human-led control efforts.
To determine if a post on StackOverflow is generated by ChatGPT, you could look for certain characteristics that are typical of posts generated by the model. For example, ChatGPT posts may contain repetitive or nonsensical phrases, lack coherence or context, or have a distinctive "machine-like" writing style. Additionally, you can check the user profile of the poster to see if it indicates that the user is a language…
okay, I did not realise that was GPT until the second last line, GPT seems to have a tendency to "hedge" all it's opinion in a certain way—it very rarely says one thing without adding a "however", "some people" etc.
Re: Use of ChatGPT generated text for posts on Stack Overflow is temporarily banned
#84Earlier quoted context omitted.
Here's one: https://stackoverflow.com/a/74675088/19877232
I don't know react. Is this correct? It looks correct. And a good example of how to use ChatGPT - a human asks for the answer and then augments it with their expertise.
The post itself contains a correction, that the example code is out-of-date and incorrect. The human answer is valid, but would be valid with or without the assistance. The assistance requires constant supervision by a subject-matter expert.
Re: Use of ChatGPT generated text for posts on Stack Overflow is temporarily banned
#85What is the mechanism by which a text is determined to originate from ChatGPT? This will be crucial for preventing unfavorable training loops among a myriad of other human-led control efforts.
The mechanism is if it reads like a middle school MLA formatted essay. I'd say at least 50% of the answers returned from chat GPT I've seen have ended with the phrase "in conclusion" or "overall".
Re: Use of ChatGPT generated text for posts on Stack Overflow is temporarily banned
#86Earlier quoted context omitted.
I don't think so, especially for StackOverflow. I want correct answers, not answers that just look correct.
Yep. Stable diffusion is powerful because there's no "right answer". Just better/worse variations of a prompt. GPT is going to be amazing for telling stories. Particularly, for "too literal" software engineers, like me. However, it doesn't seem to offer much in terms of hard, factual information that technical fields rely on.
Hard disagree on the storytelling abilities at least with the current capabilities I found the storytelling to be extremely predictable, linear and not really able to come up with unusual or interesting plots...
Which makes sense considering that these large scale language models represent the diluted sum of human generated content and most humans are by definition relatively mundane, as are the works that they produce.
Now I won't argue that the stories that these large scale language models can produce are highly cohesive, they're just not very interesting or compelling in any form.
Re: Use of ChatGPT generated text for posts on Stack Overflow is temporarily banned
#87Just finished Neil Stephenson's Anethem. In it (I'm paraphrasing here) the open internet is destroyed by media companies who flood the internet with "good garbage", that is stuff that is _almost_ correct. It forced people to pay for curated accurate content (that is, pay the same people who generated the good garbage in the first place).
This is the scary thing about this - I keep asking it questions, and it gives confident, lengthy responses that look perfect. But when you read it thoroughly, you start noticing small but important mistakes that are not obvious at first.
Why have this mountain of noise? It'll end in:
a) Wow such a nice tool that generates nice prose for emails and communication! b) Wow, such a nice tool that filters and summarizes the mountains of prose coming into my communication channels!
It's as if we're just making the channel broader with less information for no reason.
Re: Use of ChatGPT generated text for posts on Stack Overflow is temporarily banned
#88I tried ChatGPT with a few SO questions as input, it was completely missing the point of the questions more often than not. I got 3 bad answers and one correct one out of it. The scary part was just how confidently incorrect it was. The text looked very good, but there were big errors in there. I also tried to ask it a scientific question on a pretty niche field, and it produced a very reasonably looking answer. Most…
Scientific questions are, without being disparaging, mostly trivia. ChatGPT knows all of Wikipedia. If an answer could be found by extracting a passage from a Wikipedia article, ChatGPT will do a good job of answering it. Ask ChatGPT to do any engineering work, and it falls flat. It's poor mathematical skills and lack of ability to create complex plans mean that it is relying purely on it's coherent-sounding dialog.
Slight correction: ChatGPT will PROBABLY do a good job of answering factual questions. It still sometimes hallucinates even when asking known facts, thus it inherently can't be trusted.
Re: Use of ChatGPT generated text for posts on Stack Overflow is temporarily banned
#89Good. I'm wondering when HN will follow suit. Begun the AI war has.
Re: Use of ChatGPT generated text for posts on Stack Overflow is temporarily banned
#90Good. I'm wondering when HN will follow suit. Begun the AI war has.
I actually got emailed personally by Dan last weekend to stop AI-posting after doing this for more than a year.