2

I just saw the first question quoting ChatGPT and asking whether the quote is accurate.

Is this or should this be an ongoing feature of MY?

The ChatGPT does give references; perhaps the OP should just look them up himself??

3
  • 1
    Related: judaism.meta.stackexchange.com/questions/1823/…
    – Isaac Moses Mod
    Mar 15, 2023 at 20:38
  • Can this question be answered by a mod, please. I recently asked a question to gpt who gave me a really good sounding answer yet with no mekoros. I would love to know if I can post it to see if anyone can help me find sources... Would that be allowed? Apr 10, 2023 at 4:57
  • @fulltimekollelguy i don't think I'd close such a question, assuming its premise was honest (see my comment below). However, I would recommend that you first do some actual research. First step: ask ChatGPT what its source was for any interesting assertions. If it can't give you one that you can verify, the assertion may have been "Torah style" but not Torah. Second step: use the resources you have to search for a source yourself. After you've done all that, if you don't have an answer yet, you probably at least have the makings of a worthwhile question.
    – Isaac Moses Mod
    May 2, 2023 at 2:29

1 Answer 1

-1

My opinion would be that this type of question didn't necessarily have to have been closed, but rather someone should have answered "no" and then all future questions should be closed and pointed to it.

However, it and all further questions like it being closed also makes sense.

2
  • 1
    Had the post in question been honest about the fact that the response it quoted was the result of outright asking ChatGPT to lie, or, better yet, had it not come from such a request, it would probably have stayed open, and an answer could have been posted indicating that the particular citations in question are fake. However, that wouldn't indicate whether other sources that people might find in ChatGPT output in the future are necessarily also fake. More importantly, given the pace of AI progress, it certainly wouldn't indicate anything about sources cited in near-future AI outputs.
    – Isaac Moses Mod
    Mar 27, 2023 at 16:05
  • @IsaacMoses I'm sold, my only regret is I can't downvote my own answer
    – Rabbi Kaii
    May 9, 2023 at 8:35

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .