Dec 06, 2022 11:39:00

Conversational AI 'ChatGPT' Answer Posting Temporarily Banned on Coding Q&A Site Stack Overflow

Posts using ChatGPT were temporarily banned in response to the rapid increase in responses using the language model for dialogue, '

ChatGPT ', on Stack Overflow , a knowledge community about programming technology.

Temporary policy: ChatGPT is banned - Meta Stack Overflow
https://meta.stackoverflow.com/questions/421831/temporary-policy-chatgpt-is-banned

ChatGPT, developed by OpenAI, is not only capable of natural text-based conversation, but is also very good at coding, and it has been reported that it was able to pass a college-level descriptive exam.

Dialogue AI ``ChatGPT'' passes the free description problem of college student level exam - GIGAZINE

Meanwhile, Stack Overflow announced on December 6, 2022 that it will temporarily ban posting using ChatGPT. The biggest problem with Stack Overflow moderators is that ``even though the answer generated by ChatGPT has a low correct answer rate, it is easy to create a good answer at first glance''.

ChatGPT does not require specialized knowledge, so anyone can easily answer questions and output code that can be used as a model, but in fact it is not possible to judge that it is inaccurate unless you have specialized knowledge. requires scrutiny by But since Stack Overflow is powered by the goodwill of volunteers, the influx of thousands of ChatGPT-generated posts has overwhelmed Stack Overflow's foundation for maintaining the quality of questions and answers.

For this reason, ChatGPT moderators have temporarily banned the use of ChatGPT to create posts on Stack Overflow. also announced penalties. Moderators point out that this measure is temporary to curb the flood of responses by ChatGPT, and that the final decision should be decided by Stack Overflow staff and users themselves.

In response to Stack Overflow's announcement, a user said, 'Great! I sincerely

hope that ”and ``Honestly, I think that the answer by AI should be called plagiarism. I'm posting it,' and praised opinions were received.

On the other hand, 'It's all good that AI answers are not allowed, but what can we do as curators? As moderators point out, AI answers can sometimes look like good answers , a skilled person might find it wrong and downvote it, but that wouldn't be a valid reason to flag it.' There was also a comment asking whether we should identify and respond to it, and ``I completely agree with using a bot to answer.If it can be automated, it would be silly not to do it.'' We also saw opinions .

In addition, one half-jokingly asked ChatGPT the question, 'Why should AI-generated answers be banned on Stack Overflow?' Because it is not there, it may confuse or mislead users who are looking for help on Stack Overflow.' Yes, and got a lot of upvotes.