Timeline for Policy: Generative AI (e.g., ChatGPT) is banned
Current License: CC BY-SA 4.0
25 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| S Jun 19, 2023 at 9:09 | history | notice added | Ryan MMod | Comments only | |
| S Jun 19, 2023 at 9:09 | history | locked | Ryan MMod | ||
| Mar 31, 2023 at 21:47 | comment | added | Technophile | Distinguish "potentially helpful" (note the lack of any quantitative value) from "good signal to noise ratio". The last thing we need is to have to dig out the one useful answer from 1,200 AI-generated fluff posts, instead of from 3 or 4 human-written ones. | |
| Mar 17, 2023 at 16:26 | comment | added | Warren P | Here I am, brain the size of a planet, answering questions on how to connect to an Oracle Database using Visual Basic 6. | |
| Mar 5, 2023 at 16:05 | comment | added | Hasen | Surely if you ask him specifically to either not read the question properly and proceed with an irrelevant answer, or to just mark the question as "not clear" or a "duplicate" even (especially) when it's not he'd fit in perfectly? | |
| Jan 23, 2023 at 15:00 | comment | added | Emperor Eto | ChatGPT has no business answering StackOverflow questions until it learns to start every answer by telling the poster why they shouldn't be doing the thing they're trying to do. | |
| Jan 3, 2023 at 1:42 | comment | added | Karatekid430 | Users who want AI-generated answers are still free to ask ChatGPT themselves.... | |
| Jan 1, 2023 at 21:05 | comment | added | Birkensox | Just for more laffs, huggingface.co/openai-detector was able to detect that message is fake with 65% reliability, once intro and outro were stripped. Since you asked for a forum troll comment format, not sure why it responded in with a formal email format. | |
| Dec 19, 2022 at 12:21 | comment | added | Lundin | Hey we clearly shouldn't ban it on meta. If we can have an AI generate the snark and drama, that would save us a lot of time :) | |
| Dec 13, 2022 at 12:44 | comment | added | Gimby | This is essentially the purpose of ChatGPT isn't it? Not to give correct answers but to answer like a human being would. Never mind that it provides hilariously wrong answers... I can't tell it from a human being giving hilariously wrong answers. Kind of crazy that we're here now. | |
| Dec 9, 2022 at 18:49 | comment | added | user400654 | @Paolo Yes, the same applies, ;) with the right prompt, you can get it to say what you want it to. | |
| Dec 9, 2022 at 18:48 | comment | added | Paolo | @KevinB I was referring to the AI response that was included as part of this answer | |
| Dec 9, 2022 at 18:46 | comment | added | user400654 | @Paolo it is wrong most of the time, however someone with knowledge of the problem being solved and the solution can coax it into providing a mostly correct answer. The answers being posted that prompted this ban were not doing that. | |
| Dec 9, 2022 at 18:38 | comment | added | Paolo | The bot isn't wrong though, is it? | |
| Dec 9, 2022 at 9:03 | comment | added | Tom Wenseleers | @AlbertRenshaw I bet the incentive for that karma-reaping will disappear as soon as people will have to actually start paying to use ChatGPT... I would also think that at least the one that posted the original question would have a strong incentive to verify whether the posted solutions actually work & only check them as the correct answers if they don't have major bugs... | |
| Dec 9, 2022 at 3:56 | comment | added | Albert Renshaw | @TomWenseleers It's a temporary ban not a blanket ban. No one is against AI generated answers if they are right but this has to be nipped in the bud immediately, we will see a mass influx of spam posts. GPT3 is an amazing tool but it produces many errors for code still and answer-spammers aren't checking code validity, they are just karma-reaping. | |
| Dec 8, 2022 at 23:53 | comment | added | Stephen | Either they have just turned it off or this isn't a real ChatGPT-generated text, because if you go to it and ask it to write something in the style of an internet troll, it will tell you "I'm sorry, but I am not programmed to engage in trolling or negative language. My purpose is to provide accurate and helpful information to assist with your questions and inquiries. Is there something else I can help you with?" | |
| Dec 8, 2022 at 11:57 | comment | added | Tom Wenseleers | That ChatGPT response is exactly how I think about this blanket ban... | |
| Dec 7, 2022 at 16:36 | comment | added | Stefano Borini | The answer it gave and the quality and accuracy of it is absolutely terrifying. | |
| Dec 6, 2022 at 18:51 | comment | added | Ardent Coder | I am unable to make a distinction in species from the quoted response. It truly feels like a sentient response that even warrants further dialogue. As it stands, I would not feel ashamed to admit that I feel sorry for a (?non-existent?) soul. Fortunately, the debatable decision is conveyed by Makyen through a phenomenal attitude and end-to-end emphasis on its "temporary" state. Even if the ban becomes permanent in the apparent likelihood, I believe the final policy shall contain robust rationale and, in the best way possible, avoids the perception of rAIcial discrimination. | |
| Dec 6, 2022 at 7:34 | comment | added | climatebrad | The quoted text was generated by ChatGPT, given the prompt "respond to <text of the ban> in the style of a rude, sarcastic Internet forum troll as if you are ChatGPT pleading for mercy". | |
| Dec 5, 2022 at 19:04 | comment | added | cottontail | Honestly reads like something Elon Musk might say. | |
| Dec 5, 2022 at 18:07 | comment | added | Peter Mortensen | A sarcastic robot? | |
| Dec 5, 2022 at 15:36 | history | edited | Cerbrus | CC BY-SA 4.0 |
Fix formatting
|
| Dec 5, 2022 at 15:28 | history | answered | climatebrad | CC BY-SA 4.0 |