Timeline for AI-generated Answers experiment on Stack Exchange sites that volunteered to participate
Current License: CC BY-SA 4.0
Post Revisions
126 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Sep 6, 2025 at 2:05 | comment | added | redoc | Just be yourself, everyone else is already taken | |
| Apr 29, 2025 at 19:40 | history | edited | V2Blast | CC BY-SA 4.0 |
clarified phrasing
|
| Apr 29, 2025 at 15:37 | history | edited | Cesar MStaffMod | CC BY-SA 4.0 |
added 179 characters in body
|
| Apr 8, 2025 at 9:15 | comment | added | Mari-Lou A Слава Україні | @NoDataDumpNoContribution perhaps the experiment is being carried out by Google search engine as we speak. Each and every time I google for the definition of a word, or search if a phrase is idiomatic I get a bloody ai response. Is there a way to disable this feature without knowing something about programming. | |
| Apr 8, 2025 at 6:16 | comment | added | NoDataDumpNoContribution | I wonder why there are no regular updates of experiments or even a suggested duration at the beginning. Is the experiment still running? Is the answer bot still answering? If yes, when will conclusions be available? | |
| Mar 4, 2025 at 16:55 | history | edited | SashaStaffMod |
edited tags
|
|
| Mar 4, 2025 at 16:49 | history | edited | Starship | CC BY-SA 4.0 |
typo
|
| Feb 24, 2025 at 16:39 | answer | added | N. Virgo | timeline score: 11 | |
| Feb 20, 2025 at 19:31 | answer | added | Therac | timeline score: 13 | |
| Feb 20, 2025 at 2:23 | answer | added | cubick | timeline score: 9 | |
| Feb 19, 2025 at 18:15 | comment | added | JDB | The AI issue is already covered well... but that example! It's a terrible example! "best brewing method"??!? It's blatantly off-topic and exactly the kind of question we don't want. How about an AI that can detect and flag potentially bad questions, rather than trying to answer them. | |
| Feb 18, 2025 at 9:52 | comment | added | NoDataDumpNoContribution | @manassehkatz-Moving2Codidact On the sites where this experiment runs (if it still runs, hopefully not) AI generated content is allowed. The ban isn't network-wide. | |
| Feb 16, 2025 at 16:44 | comment | added | manassehkatz-Moving 2 Codidact | @NoDataDumpNoContribution Absolutely impossible to do a double-blind experiment here, because the frequent users are used to downvoting AI hallucinations, so rather than improve answers that are from AI but not displayed as being from AI, they will figure it out and say "some idiot posted AI junk here, get rid of it!", not knowing that "some idiot" was actually a SE-sanctioned experiment. | |
| Feb 15, 2025 at 17:02 | comment | added | NoDataDumpNoContribution | Is there a timeline how long the experiment will run approximately? I'm also curious about the stats afterwards. Like how the bot answers fared. To be completely fair one in principle would need to performa double blind experiment, but I guess that would require too much trust. Also, if the results are too disastrous, maybe we can just stop AI experiments for 1-2 years and go back to the drawing board. | |
| Feb 14, 2025 at 15:08 | answer | added | EJoshuaS - Stand with Ukraine | timeline score: 6 | |
| Feb 14, 2025 at 10:13 | comment | added | NoDataDumpNoContribution | The other day I had the thought if only more subject matter experts would be working at SE, they would surely see how inadequate LLM based bot answers are. We wouldn't need to conduct the experiment (and put resources into developing infrastructure for it) to get an insight that is kind of trivial for a subject matter expert. However, an experiment that is fairly evaluated is in principle not the worst. We will learn something for sure we only (strongly) guessed before. However, it's very important it's not misused and only performed pro forma. | |
| Feb 13, 2025 at 4:30 | answer | added | Ignis Incendio | timeline score: -15 | |
| Feb 12, 2025 at 16:21 | comment | added | deee | @gerrit the example answer is a great example of something that is more or less correct so will get marked as "correct" by people not paying much attention, and also a completely useless answer to the question actually asked | |
| Feb 12, 2025 at 9:48 | comment | added | Ander Biguri | I assume you have done a very thorough market research right? Because as someone who teaches computing courses at Cambridge, I can tell you one certain thing: all my students that use AI for programming questions directly ask chatGPT, they don't even know SO. Only the ones interested in human answers come here. Are you genuinely thinking they will now get off their standard platform to come here? This has a "I am going to make a new facebook, but better!" vibes. | |
| Feb 11, 2025 at 16:18 | answer | added | Agent_L | timeline score: 11 | |
| Feb 11, 2025 at 11:57 | comment | added | NoDataDumpNoContribution | @AJM Typically I would say that it stands for them not only saying it, but also doing it. But maybe it also stands for them only wanting it if providers can provide it and otherwise not wanting it. Probably depends all on what the definition of non-negotiable is. | |
| Feb 11, 2025 at 11:54 | comment | added | NoDataDumpNoContribution | @Berthold Can I ping you again with the comment asking about the license of the content provided by the bot? The excitement about the experiment has gone done a bit so it may be time to concentrate more on the fine print, like what are people actually editing there. There is a whole question about that. | |
| Feb 11, 2025 at 10:23 | answer | added | CharonX | timeline score: 14 | |
| Feb 11, 2025 at 10:07 | comment | added | AJM | "We’ve stated that attribution is non-negotiable and that goes both ways" What does "that goes both ways" mean here? | |
| Feb 11, 2025 at 9:45 | answer | added | kwypston | timeline score: 12 | |
| Feb 11, 2025 at 9:44 | answer | added | LWChris | timeline score: 10 | |
| Feb 11, 2025 at 9:20 | comment | added | gerrit | Related: Partnership with Google Gemini, Partnership with OpenAI | |
| Feb 11, 2025 at 9:00 | comment | added | tenfour | When this gets implemented despite all the feedback in this thread, think hard about how SE as an organization values its most active and eager contributors. | |
| Feb 11, 2025 at 8:48 | comment | added | gerrit | Did you select a terrible answer on purpose? | |
| Feb 11, 2025 at 8:28 | answer | added | Mr. Squirrel.Downy | timeline score: 8 | |
| Feb 10, 2025 at 22:36 | comment | added | NoDataDumpNoContribution | With the kind of perfect string of AI experiment failures: AI search never took off, AI reformating was a complete disaster, now AI answers that are garbage, I wonder why it's so difficult to introduce some sort of AI here. Even users propose some things (duplicate finder, comment summaries) but they definitely aren't done while the things that are done are useless. One cannot be so unlucky. There must be some method behind, which is puzzling. | |
| Feb 10, 2025 at 22:31 | comment | added | NoDataDumpNoContribution | @Seth "Why would I come to this website if I have to help a machine, not a person?" You might get paid or for the greater good or because you do not value personal interaction that much maybe? Don't want to disturb just give some potential reasons why somebody might actually want to do that. | |
| Feb 10, 2025 at 20:07 | comment | added | Seth | StackOverflow et. al. is a "Q&A Website", not an "A" website. The point is to build a creative commons, where people help people. The entire process described here is an acknowledgment that LLMs don't work. Why would I come to this website if I have to help a machine, not a person? And it's not like I'd even helping the machine learn and get better, I'd be literally cleaning up its garbage! You're asking all the contributors to SE sites to pick up machine-generated garbage. The machines should be picking up OUR garbage, not the other way around. | |
| Feb 10, 2025 at 13:24 | comment | added | Ramhound | The last person who should be determining if an AI answer is correct, is the author of the question, since they don’t have enough domain knowledge to answer the question (most of the time). When they do have enough domain expertise they won’t be verifying an AI generated answer. So once again the team has failed the community. | |
| Feb 9, 2025 at 12:24 | comment | added | Vincent Thacker | "Committed to the Stack Exchange network being a place for human-curated knowledge and information" With all due respect, this is nothing more than meaningless corpo-speak. | |
| Feb 9, 2025 at 0:27 | comment | added | cfr | I was asked to explain my downvote: If attribution is non-negotiable, it is a fundamental criterion any answer must satisfy if it is to survive scrutiny. To make that claim in one breath and then blithely announce you're deliberately adding answers which fail to satisfy that criterion makes no sense at all. And as a reminder, these answers more than any are necessarily somebody else's content. A human answer may cite no source because it is the work of the answerer. The same cannot be said for so-called AI. | |
| Feb 8, 2025 at 21:14 | answer | added | John Omielan | timeline score: 10 | |
| Feb 8, 2025 at 13:34 | answer | added | Karl Knechtel | timeline score: 25 | |
| Feb 8, 2025 at 6:10 | comment | added | galacticninja | Related: Why was Web Apps SE excluded from the current stage of the AI "Answer Bot" experiment? | |
| Feb 8, 2025 at 6:05 | answer | added | cottontail | timeline score: 9 | |
| Feb 7, 2025 at 19:14 | answer | added | BMitch | timeline score: 13 | |
| Feb 7, 2025 at 14:18 | answer | added | ColleenV | timeline score: 7 | |
| Feb 7, 2025 at 14:15 | comment | added | Tim Lewis | To me, the most frustrating thing about seeing AI answers on Stack sites is the thought "If the asker wanted an AI-generated answer, they would have asked an AI." Adding this "Answer Assistant" will change that thought to "Why bother with asking a question on <StackSite> at all if an AI is just going to answer it anyway? Just ask an AI, skip the middleman." I realize you want these answers "curated by the community", but I have negative interest in doing that... I like helping people, not AI 🫠 | |
| Feb 7, 2025 at 8:10 | answer | added | Ale | timeline score: 27 | |
| Feb 7, 2025 at 0:59 | answer | added | Mark | timeline score: 24 | |
| Feb 6, 2025 at 23:16 | comment | added | DBS | Generally, the people qualified to properly proofread an LLM generated answer are already the ones answering questions. This seems like it will just encourage people who have a moderate understanding of the topic to mark the AI as correct if the output "looks about right" (which it usually does, creating text that seems relevant is what LLMs do best) TL;DR: I am willing to help humans solve problems, I am not willing to proofread LLM guesswork. | |
| Feb 6, 2025 at 22:16 | answer | added | Jan Murphy | timeline score: 30 | |
| Feb 6, 2025 at 19:37 | answer | added | NotThatGuy | timeline score: 7 | |
| Feb 6, 2025 at 17:43 | answer | added | kaya3 | timeline score: 13 | |
| Feb 6, 2025 at 13:39 | answer | added | Mari-Lou A Слава Україні | timeline score: 26 | |
| Feb 6, 2025 at 13:34 | answer | added | JRE | timeline score: 24 | |
| Feb 6, 2025 at 8:50 | comment | added | Lundin | SO will be one of the companies we should blame for contributing to that hype. If they will be around at that point and not go bankrupt when the bubble bursts. Inflated and overhyped Nvidia stock is already plummeting, the bursting has started. | |
| Feb 6, 2025 at 8:48 | comment | added | Lundin | @Ray It is exactly the same hype as the Internet hype in the early 2000s. Like AI, Internet was a technology there to stay, but buzzword market people started to create an inflated hype claiming that Internet was the solution to every problem in the world, much like snake oil. The core technology was still there to stay, but the hype caused a massive crash when the bubble burst. Yes, AI companies will suffer, general tech companies too and probably non-related companies. People will lose their jobs. | |
| Feb 6, 2025 at 2:56 | answer | added | ApexPolenta | timeline score: 18 | |
| Feb 6, 2025 at 1:31 | comment | added | Excorpion | Isnt that the point of having us to filter the info? Now we have to make sure users have good answer, but also that the IA have good answers too ? | |
| Feb 6, 2025 at 1:06 | answer | added | Cris Luengo | timeline score: 41 | |
| Feb 6, 2025 at 1:05 | answer | added | anongoodnurse | timeline score: 66 | |
| Feb 5, 2025 at 21:56 | comment | added | Berthold StaffMod | @ТymaGaidash To limit the exposure of answers that include incorrect information | |
| Feb 5, 2025 at 21:40 | answer | added | Weird Glyphs | timeline score: 24 | |
| Feb 5, 2025 at 19:03 | answer | added | Shadur-don't-feed-the-AI | timeline score: 59 | |
| Feb 5, 2025 at 18:30 | answer | added | Josh Brunton | timeline score: 16 | |
| Feb 5, 2025 at 18:10 | comment | added | Tyma Gaidash | Why not have the deleted answers be visible to 10k users as well? | |
| Feb 5, 2025 at 18:02 | comment | added | Ray | @Lundin "All who remain in the hype are losers" The other group that stands to lose are the actual scientists. AI isn't snake oil; there are a lot genuinely useful tools coming out of AI research, all of which have both advantages and limitations. The problem is that someone taught the phrase "AI" to business majors and now they're selling it as a magic panacea to all problems and don't care whether it can do what they claim as long as they can make a quick buck. When the hype collapses, research funding will likely suffer. It's happened before. | |
| Feb 5, 2025 at 17:59 | answer | added | ColleenV | timeline score: 28 | |
| Feb 5, 2025 at 17:33 | answer | added | Francisco IA Lover | timeline score: 6 | |
| Feb 5, 2025 at 17:11 | comment | added | java-addict301 | We've already got AI generated answers - it's called Google. If you try searching any tech question now, it pops right up. SO is quickly losing its relevance, and I say this as someone who has never publicly criticized the company before. As a software engineer, I only use it about 50% of the time that I used it even just a few months ago due to this Google feature, and I will certainly be using it much less now. | |
| Feb 5, 2025 at 16:03 | comment | added | Joe W | Is there any way to get a list of all the answers this user has provided? I haven't been able to see any of them on web apps or raspberry pi. | |
| Feb 5, 2025 at 15:21 | answer | added | Resistance Is Futile | timeline score: 68 | |
| Feb 5, 2025 at 13:51 | answer | added | Ian Kemp - SE killed by LLMs | timeline score: 59 | |
| Feb 5, 2025 at 13:50 | comment | added | SPArcheon | @Lundin I know. That is why in my answer I say that I am pretty sure the discussion was done with a few mods that they knew would agree (remember, there is a precedent for "we told some cherry picked users that are more equal than the others about this in advance") | |
| Feb 5, 2025 at 13:43 | comment | added | Lundin | @ꓢPArcheon I just went through the metas on the 3 sites that "volunteered" and found no prior discussion on any of the sites, just announcements: "we are going to do this experiment now". I don't think there was any public discussion at all. Getting the majority of all moderators to agree on a small site with just 1 or 2 active moderators won't be that hard. | |
| Feb 5, 2025 at 13:37 | comment | added | SPArcheon | @AJM While I totally expected unreviewed AI generated answers to still be counted and thus skewing the "Answers:0" filter... they apparently aren't so at least this feature hasn't been broken. Yet. | |
| Feb 5, 2025 at 13:04 | comment | added | SPArcheon | @Jeremy "In my experience, when you’re using a system that tries to provide attribution, it hallucinates answers and then slaps on a tangential attribution that doesn’t actually say what’s claimed, unless it’s operating in a very narrow scope like an internal knowledge base".... good. And now, what does this tell you about the maturity of this tech and trying to shoehorn it into something that has attribution as its basic requirement? Wouldn't it be better to research other alternatives that actually fits the role? | |
| Feb 5, 2025 at 12:59 | comment | added | SPArcheon | "Several Stack Exchange communities have volunteered" -citation needed. Either the community agreed and please point us to the public vote pools, or you got some mods to agree as you said before. It can't be both. You may like thinking that, but a few mods that are friend to the company do not represent a community as a whole. They probably don't even represent all the mods. Please call things as they are. What you did is getting a majority of the local mods of some sites to agree with you, which may not be enough for some users given how utterly protective some mods are about the company... | |
| Feb 5, 2025 at 11:10 | comment | added | AJM |
Will there be a way for someone asking a question to opt out of AI answers to it? How will this affect searches based on answers:0 by people looking for unanswered questions to answer?
|
|
| Feb 5, 2025 at 11:09 | history | edited | Journeyman Geek | CC BY-SA 4.0 |
Fixed spelling error
|
| Feb 5, 2025 at 11:02 | answer | added | Lundin | timeline score: 113 | |
| Feb 5, 2025 at 10:03 | answer | added | SPArcheon | timeline score: 22 | |
| Feb 5, 2025 at 9:59 | comment | added | NoDataDumpNoContribution | @Berthold It would be very interesting to have information about the license of the generated content. What is it licensed under? Is it compatible with CC-BY-SA? Could this information maybe be added to the question? | |
| Feb 5, 2025 at 9:54 | comment | added | l4mpi | You call it an experiment, I call it shooting yourself in the foot... but that seems to be all the rage in the US right now, looks like SE was worried it might miss that train. Let's just say that if this ever comes to SO I hope the community will actively sabotage this "feature" - thankfully at the moment the way it is built requires users to approve the answers, so as long as there are enough users who disapprove of the AI slop the community would have ways to protest against this. | |
| Feb 5, 2025 at 9:39 | comment | added | Sayse | Just call a spade a spade. The last time you did an experiment, it magically just became a new feature with some random unexplained marginal growth metric | |
| Feb 5, 2025 at 9:30 | answer | added | Lomtrur | timeline score: 26 | |
| Feb 5, 2025 at 8:54 | answer | added | Steve Bennett | timeline score: 23 | |
| Feb 5, 2025 at 7:59 | comment | added | Lundin | The cure of tulip mania is not to eat tulips for breakfast, lunch and dinner. The cure is the realization that you are spending a fortune on useless and likely unhealthy tulips! Ask Nvidia how the AI hype is working out for them. Those who made profit on the AI hype have already cashed out now. All who remain in the hype are losers and it's just a matter of how much you will lose from now on. You are going to arrive to the market with your wagons of tulips to realize nobody will pay $100 for a tulip when they can pay $1 for a loaf of bread. | |
| Feb 5, 2025 at 4:12 | answer | added | starballMod | timeline score: 14 | |
| Feb 5, 2025 at 2:16 | answer | added | PM 2Ring | timeline score: 37 | |
| Feb 5, 2025 at 0:52 | answer | added | M-- | timeline score: 17 | |
| Feb 5, 2025 at 0:05 | answer | added | Adamant | timeline score: 32 | |
| Feb 4, 2025 at 23:49 | comment | added | security_paranoid | While a lot of people are unhappy about this, I do appreciate you taking the time to openly let us know what is happening, even if we get little say in it. | |
| Feb 4, 2025 at 23:47 | answer | added | security_paranoid | timeline score: 5 | |
| Feb 4, 2025 at 23:33 | comment | added | JBH | What's the point of Stack Exchange if AI-generated answers are permissible? Querents could just type the question into Google and get the same (theoretically) result. Humans have an intuition that AI may never have, making the humans the only valuable contributors to Stack Exchange. ... Unless your goal is to be rid of the pesky humans.... | |
| Feb 4, 2025 at 22:45 | comment | added | Joe W | From what I understand the different test sites have different reputation required to see these answers so it would be good to list what the required rep for each site is. | |
| Feb 4, 2025 at 22:41 | comment | added | Kaia | I've tracked down some sample AI answers on the Arts and Crafts site: crafts.stackexchange.com/q/13320 crafts.stackexchange.com/q/13311 crafts.stackexchange.com/q/13267 crafts.stackexchange.com/q/13254 | |
| Feb 4, 2025 at 22:30 | comment | added | NoDataDumpNoContribution | I think this undermines the GenAI ban on some SE sites. As long as for example the GenAI ban lasts on SO, this answer generator would have no chance there. But when it is tested now, it may be introduced nevertheless. Wouldn't this again result in a moderator strike? If we do not pay attention, we might end up with more AI answers than human answers. A competition of AI generated answers and human generated answers will only result in alienating human answerers even more. This might not end well. AI and human content must be extremely well separated if AI content is to used at all. | |
| Feb 4, 2025 at 22:26 | history | edited | BertholdStaffMod | CC BY-SA 4.0 |
Not sure what this was meant to add, please don't edit the summary.
|
| Feb 4, 2025 at 22:21 | answer | added | NoDataDumpNoContribution | timeline score: 52 | |
| Feb 4, 2025 at 22:13 | comment | added | Ray | "The current experiment is integrated with an existing data partner...We are not able to disclose the specifics during this phase of the experiment." - Sounds legit. | |
| Feb 4, 2025 at 22:00 | answer | added | Franck Dernoncourt | timeline score: 29 | |
| Feb 4, 2025 at 21:56 | history | edited | Franck Dernoncourt | CC BY-SA 4.0 |
typos
|
| Feb 4, 2025 at 21:54 | answer | added | Franck Dernoncourt | timeline score: 3 | |
| Feb 4, 2025 at 21:46 | answer | added | Andras Deak -- Слава Україні | timeline score: 155 | |
| Feb 4, 2025 at 21:19 | answer | added | Draconis | timeline score: 125 | |
| Feb 4, 2025 at 20:48 | answer | added | Kaia | timeline score: 63 | |
| Feb 4, 2025 at 20:26 | answer | added | Kaia | timeline score: 48 | |
| Feb 4, 2025 at 20:00 | comment | added | Berthold StaffMod | @hkotsubo The association bonus is indeed something to consider, and UX actually has the rep requirement set to 151 for that reason. We'll also be tracking whether users who interact with the experiment are able to see it by virtue of the bonus. | |
| Feb 4, 2025 at 19:39 | comment | added | Resistance Is Futile | "Committed to the Stack Exchange network being a place for human-curated knowledge and information." No. It is a place for human created and curated knowledge. You will have hard time finding experts willing to curate AI nonsense. | |
| Feb 4, 2025 at 19:15 | answer | added | TylerH | timeline score: 42 | |
| Feb 4, 2025 at 19:12 | history | edited | TylerH | CC BY-SA 4.0 |
reworded the title to clarify what this is about
|
| Feb 4, 2025 at 19:11 | answer | added | Joe W | timeline score: 51 | |
| Feb 4, 2025 at 19:11 | comment | added | hkotsubo | "Users with at least 50+ rep (default value) on the specific Stack Exchange site will be able to see and evaluate the private answers." - Currently, the association bonus is 100 rep, which means that I could join Arts & Crafts, Raspberry Pi or User Experience (UX), earn the bonus and immediately start reviewing answers without any prior participation on those sites. Wouldn't it be better to increase this threshold? Probably the same rep required to access the review queues (which at least requires some participation on my part before I can review anything). | |
| Feb 4, 2025 at 19:02 | answer | added | John Omielan | timeline score: 25 | |
| Feb 4, 2025 at 18:46 | answer | added | Dharman | timeline score: 185 | |
| Feb 4, 2025 at 18:43 | comment | added | user400654 | I don't quite understand the name of this experiment... This isn't an answer assistant, it isn't assisting anyone at answering a question... instead it's an attempt at filling in for the lack of answerers. | |
| Feb 4, 2025 at 18:41 | answer | added | user152859 | timeline score: 13 | |
| Feb 4, 2025 at 18:39 | comment | added | NotThatGuy | We need fewer, higher-quality Q&As - AI could be useful as a personal assistant, to provide personalised responses, but it doesn't belong in public posts. But I guess that idea isn't as profitable as just scaling up volume at the expense of quality. | |
| Feb 4, 2025 at 18:29 | answer | added | Thomas Owens | timeline score: 245 | |
| Feb 4, 2025 at 18:24 | answer | added | Starship | timeline score: 227 | |
| Feb 4, 2025 at 18:21 | comment | added | user400654 | Does seem a bit disingenuous to claim so much intent to have properly cited answers... when stack hasn't even proven it's possible, while moving forward with actually publishing ai generated content without it. Given how long it takes for stack to iterate on things now days it could be years before this ever gains any form of citation feature, long after far too much of it has been posted to reasonably clean up the mess. | |
| Feb 4, 2025 at 18:14 | answer | added | user400654 | timeline score: 8 | |
| Feb 4, 2025 at 18:05 | comment | added | user400654 | "We want to test if this feature could help improve the answer experience and encourage knowledge sharing by helping users get unstuck or get a jump-start on content curation while maintaining quality." did you mean asking process? i can only imagine this will make the answering process worse. | |
| Feb 4, 2025 at 18:01 | comment | added | user1114 | If you can push on major AI providers to improve the attribution capabilities of their output, that would be a wonderful thing for the common good. I’m skeptical, but if it is possible, it will probably require a significant change to the type of results they’re generating that would potentially skew the results of your experiment. In my experience, when you’re using a system that tries to provide attribution, it hallucinates answers and then slaps on a tangential attribution that doesn’t actually say what’s claimed, unless it’s operating in a very narrow scope like an internal knowledge base. | |
| Feb 4, 2025 at 18:00 | history | edited | cocomac | CC BY-SA 4.0 |
deleted 1 character in body
|
| Feb 4, 2025 at 17:56 | answer | added | cocomac | timeline score: 19 | |
| Feb 4, 2025 at 17:54 | comment | added | Mast | Attribution/citation/sourcing is not negotiable, but it is not included in the current tests. That seems like it's going to skew the results quite a bit. | |
| Feb 4, 2025 at 17:53 | history | edited | bobble |
edited tags
|
|
| Feb 4, 2025 at 17:38 | history | asked | BertholdStaffMod | CC BY-SA 4.0 |