Chatbots (and the large language models that power them) have become very popular; they can often output material much faster than humans can. Unfortunately, they cannot write as well as humans (yet), and their output is prone to hallucinations, false citations, and other errors. This has created a major cleanup burden at Wikipedia, as many editors (especially new ones) try their hand at using artificial intelligence to edit Wikipedia. You can help by identifying AI-written text, removing unsourced or inaccurate claims, and by identifying AI-generated images. For more information, see our AI Cleanup Guide.
16 Jul 2026 –Nemotron(talk·edit·hist) was PRODed by Newslinger (t·c): Presumptive removal of LLM-generated content, see {{slink|Wikipedia:AI noticeboard/Archive 7#Esculenta and GAs}}. Feel free to reinstate by following the procedures for disputing presumptive removal of LLM-generated content.
Comment: for what it's worth, the possibly LLM-generated text was added in 2023 (when LLMs were just starting to become prevalent and we didn't really have much policy on the matter yet), while the remaining stub has been around since 2008. (No opinion on anything else.) WCQuidditch☎✎00:48, 22 July 2026 (UTC)[reply]
I could not find any coverage for this software in WP:RS for it to meet the WP:GNG. Additionally, the page appears to have been partially or mostly generated by AI (see the weirdly named heading "Name origin and confusion" and the features list with excessive bold). There was a previous AFD at this title in 2008 but it seems to have been for something completely different. Schützenpanzer(Talk)00:37, 22 July 2026 (UTC)[reply]
Personally, I don't see any strong signs of AI-generation, especially given its first revision contains many human-like errors. AI-polished, maybe, but not wholesale generation. Catalk to me!15:40, 21 July 2026 (UTC)[reply]
Entirely LLM-generated article. The most glaringly obvious signs are the overly dramatic headings and subheadings (f.ex. The Simultaneous Epochal Matrix ('Past Forward'), The 'Blind Review', Conceptual Migration to the Stuttgart State Opera (2026)) and the bolded WP:AILISTS.
Unfortunately, I had first draftified the page but the user DRAFTOBJECTED before being blocked by copypaste moving it, so now we have to go through AfD instead. ‑‑gurkubondinn15:20, 20 July 2026 (UTC)[reply]
Redirect as suggested by the nominator. Though I am not sure there is anything worth saving here, so delete would alternatively be merited. ChrysGalley (talk) 08:46, 20 July 2026 (UTC)[reply]
KEEP. This page needs to be cleaned and fixed of the LLM-generated content, but surely not deleted. The there is page which deals with this topic currently but it cannot accommodate this. Instead, ethnic relations in India should have a link to this page as the main page in the section noted. Astra Travasso (talk) 11:12, 20 July 2026 (UTC)[reply]
Delete/Redirect LLM generated misleading content. Also agree with the nom that the topic enough does not really warrant to have its own article. THEZDRX(User) | (Contact)15:25, 20 July 2026 (UTC)[reply]
Delete/Redirect. I have carried out the kind of cleanup that would be needed here a couple of times, and have reluctantly concluded that it shouldn't be done as it is more work, and riskier, than writing a new article from scratch. No prejudice to clean recreation. M kuhner (talk) 15:46, 20 July 2026 (UTC)[reply]
Delete/Redirect per nom. The article includes too much about geography and colonial times, and the sources are not explicit enough to call it a racism. There is a lot of news-related content. Whatever valuable content there is, it can be easily merged to Ethnic relations in India, which way is smaller in comparison. MapSGV (talk) 03:58, 21 July 2026 (UTC)[reply]
@X AI has created grokipedia: What article improvement is being done? It's been 30 hours since your last edit, and the only edits have been to nominate it for deletion and my edit to add categories (which won't impact this discussion). GoingBatty (talk) 21:36, 20 July 2026 (UTC)[reply]
Delete - No SIGCOV in RS. Sources in the article are primary, a business directory and listings in app stores. I didn't found any other sources. — Itzcuauhtli11 (talk) 05:29, 20 July 2026 (UTC)[reply]
Delete: An article on a start-up company/app, sourced to app store listings, etc. See also the earlier Draft:Laniakea app, with multiple AFC rejections. I am not finding evidence of attained notability. (Note there are other similarly-named companies, such as Laniakea Group Ltd.) AllyD (talk) 08:58, 20 July 2026 (UTC)[reply]
False inflation of relationship between two nations sharing a language, some cultural and colonial history which is significant in some contexts but not this one. They have only met ONCE in a competitive fixture prior to the upcoming World Cup final. Crowsus (talk) 01:51, 19 July 2026 (UTC)[reply]
Delete: I was going to draftify this article pending the events of tomorrow's match and the subsequent reporting over the next week. However, the nomination essentially summarizes the primary concern I have: there's almost nothing for this article to be about. I am also a bit concerned that this could have been written with AI, given the rapid content additions made by the creating editor. Best, ~ Pbritti (talk) 02:04, 19 July 2026 (UTC)[reply]
Delete Clearly LLM-generated, typical LLM puffery such as: Though separated by the Atlantic Ocean, the rivalry carries deep historical and cultural significance due to the colonial ties and shared language between the two nations, and is contested by two of the most successful teams in global football. –LaundryPizza03 (dc̄) 02:30, 19 July 2026 (UTC)[reply]
Keep / Comment: I respectfully disagree with the arguments for deletion.The question in this discussion should be whether the topic satisfies Wikipedia's inclusion policies, not whether individual editors personally consider the rivalry significant. Wikipedia is based on WP:V and WP:GNG, not subjective assessments of importance. Regarding the claim that this is a "false inflation" because Argentina and Spain had met only once in a competitive fixture before the 2026 FIFA World Cup Final, the number of competitive meetings is not a criterion for notability. International football rivalries are not defined solely by the number of competitive matches but by the extent to which reliable, independent sources discuss them as a distinct topic. The article is supported by multiple high-quality, independent sources from major international media that examine the history between the two national teams, the significance of their meetings, and the context surrounding the World Cup Final. That is the relevant consideration under WP:GNG. The assertion that "there's almost nothing for this article to be about" is also inaccurate. The article covers the complete head-to-head record, notable matches between the teams, the historical development of their encounters, and the significance that reliable sources attribute to the rivalry ahead of the 2026 FIFA World Cup Final. Whether editors personally agree with that characterization is secondary; Wikipedia summarizes what reliable sources publish.Regarding the suggestion that the article was generated using AI, I can state unequivocally that it was not. I began researching and drafting the article immediately after the FIFA World Cup semi-finals concluded, once the finalists had been confirmed. The article was manually researched, written, sourced, and formatted by me. The fact that it was created quickly reflects the preparation that went into it, not the use of AI. Per WP:AGF, editors should avoid assuming bad faith based solely on the speed with which an article is created. This discussion should therefore focus on the article's sourcing and compliance with Wikipedia's content policies. As it stands, the article is supported by multiple independent, reliable sources that provide significant coverage of the subject, satisfying WP:GNG. For these reasons, I believe the article should be retained. SydbourneSydbourne (talk) 02:54, 19 July 2026 (UTC)[reply]
Delete - Outside of the 2026 World Cup final, there haven't been any games to note (a World Cup group stage game in 1966). That's really it. Considering there is already a fully detailed game on this World Cup final.
Delete for all the reasons listed above. World Cup finals shouldn't have articles titled X-Y rivalry at all. This clearly contradicts the goooooool of Wikipedia. Hansen Sebastian (Talk) 08:35, 20 July 2026 (UTC)[reply]
Agreed with delete. This is more of a contrived "rivalry" with recency bias because of the 2026 World Cup. The two countries don't have much of a history of heated matches over the decades, unlike the England-Argentina rivalry that gets super passionate. We should be careful about what qualifies as a true rivalry and what doesn't. Doctorstrange617 (talk) 14:03, 20 July 2026 (UTC)[reply]
Delete, fails NOLLM. There is a lot of tortured undue emphasis pn importance and legacy, probably because of the contrived subject and the LLM attempting to predict a longer tect for the user. ‑‑gurkubondinn21:13, 20 July 2026 (UTC)[reply]
I have two main arguments as to why I believe that this topic should be deleted and then redirected back (both, not just redirected) to Minecraft modding:
I do not believe the topic is notable to begin with, with a lack of significant coverage in sources. There are some sources from the reliable PC Gamer that could be useful , but these alone are not enough to support a whole article. There's also some coverage of how Create developers were hired to work on Hytale, but that's not coverage of Create itself. Those aside, the entire article is cited to primary sources, sources of dubious reliability (e.g. games.gg), sources that can't establish notability (e.g. Valnet sources), and unreliable user-generated sites (e.g. Modrinth and the Create wiki).
I have strong reasons to believe that this article might have been AI-generated. There are many WP:AISIGNS in this article, such as awkward AI-sounding language with no human input such as "emphasizing creative freedom over prescribed solutions", the random block quote in the middle of Gameplay, the bolded pullet points with similarly promotional language ("This system has been credited with significantly reducing the mod's learning curve"), and an abundance of statements with seemingly random unrelated sources tacked onto them. Plus promotional non-factual statements without proper sourcing like "Create has become one of the most widely covered and influential mods in the Minecraft modding ecosystem, with several hundred million downloads to date". The article is either AI-generated, or promotional. And I lean towards AI-generated. From this, under WP:LLM: "the use of LLMs to generate or rewrite article content is prohibited". An article being AI-generated is a valid deletion reason (see also: WP:TNT). I probably could've nominated this for speedy deletion under WP:G15, but there does appear to be some ambiguity on if it was AI-generated and the editor has had an account for a while.
If the mod itself is actually notable, then by all means, someone could create a real article for it without this level of either AI-generated or promotional language. But what there is in the article now needs to go; see WP:TNT. λNegativeMP121:14, 18 July 2026 (UTC)[reply]
These concerns are valid. I went through and made changes addressing a lot of your concerns, but I should note that more or less everything in the article was written by me. I did reference Google AI summaries when writing, which is the reason for some of the language you point out. A lot of what you pointed out, though, did come directly from me: I am not an experienced Wikipedia editor by any means, and for that reason I wouldn't expect that everything that I write appears to follow standard Wikipedia style expectations. I also should note that I am not affiliated with the Create mod in any way, nor have I been in the past.
As for notability, I think that there is no doubt that Create is notable enough to warrant an article. You mention the PC Gamer articles as reputable, but within Gaming journalism ScreenRant, GamesRadar, and games.gg are all reputable, independent sources. Even the two articles that don't directly cover Create itself still contain a lot of information about the mod, despite focusing on the developers being hired at Hypixel Studios.
Regardless, I also added a RockPaperShotgun source directly covering Create itself, and RockPaperShotgun is well-known and well-regarded within the gaming world.
Something I should note is that, within the world of gaming, news articles aren't a very common medium. Almost all relevant coverage of the Create mod takes place in the form of YouTube or TikTok videos, because that is the kind of content that the audience for Minecraft mods consumes. For this reason, there is a disproportionately small amount of news coverage of the Create mod given its size. Realistically, the real sign of notoriety in this case are the many thousand YouTube videos about Create is the more important signal. Still, the fact that there are well over a half-dozen articles covering the Create mod is impressive, and relevant in this case because I understand that is primarily what Wikipedia considers when evaluating notoriety. The fact that it has 220 million downloads and that their discord server has well over 100,000 members is also a good sign.
Keep The sources are reliable and the article can easily be fixed. The page author directly stated that his use of AI tools was limited to referencing with what I see as clear double checking. Dabmasterars [RU/COM] (talk/contribs) 23:59, 18 July 2026 (UTC)[reply]
Redirect to Minecraft modding at least on current sourcing. I appreciate an editor has taken the time to try and amend a pretty flawed article although issues remain. Focusing on WP:GNG and WP:RS coverage about the game, I think you have a start: the two PC Gamer articles and one Rock Paper Shotgun and GamesRadar source. The sum coverage of these four is not particularly solid:
The Watts article boils to a basic description about the mod, calling it "impressive" and "ingenious", and referring the player to download it on Curseforge.
The Lane article has a useful descriptive overview of an addon to Create called Create Aeronautics. That's great, but it's describing an add-on component, not the main mod framework itself. No real evaluation of it other than calling it "huge".
The Pearson article found it "inspiring" and considered it made the game experience feel "alive and connected", but dedicates most of the article refers to what players have uploaded to YouTube using it.
Koselke is more a WP:PRIMARY interview source, with some nice dev comments. It does briefly call Create "one of the best Minecraft mods out there"; that's about it.
Whilst noting there are some other situational WP:VALNET sources here which are fine to add, the substance of the article, including technical detail on what contraptions and processes are introduced and how they work, its development background, review reception, and that it is considered "widely regarded as one of the most significant Minecraft mods of its era" is not strongly founded by the coverage in WP:RS. That suggests non-notability. VRXCES (talk) 03:43, 19 July 2026 (UTC)[reply]
I added a CBR article titled "10 Things That Make Create The Best Mod Ever". That article is a very clear evaulation of the mod and does a lot to support the claim that it is "widely regarded as one of the most significant Minecraft mods of its era". I again want to emphasize that the target audience for a Minecraft mod is not likely to read a lot of magazine articles and that this mod is relevant and notable not because of print media coverage but because of its impact. The article is well-sourced. Jerryeit (talk) 17:05, 19 July 2026 (UTC)[reply]
I respect your attempt to try and create this article, and being honest about how AI was involved, but your view of notability is not what our notability policies actually are. Reliable sources discussing a subject in detail is what makes something notable. So while you may not think that the audience of a Minecraft mod and the audience of a gaming magazine might not have much overlap, that does not matter as a need for in-depth sourcing is required by Wikipedia policy. Also, CBR is not a reliable source, per WP:VG/S, and Valnet sources can't contribute to notability anyways. λNegativeMP117:12, 19 July 2026 (UTC)[reply]
I understand all of the above, and I appreciate you explaining these things to me. Still, I am left confused: even completely disregarding all of the Valnet sources, and the other sources that you have broken down as not reliable, there are still a good 4 articles directly covering the Create mod.
Compare this to the article on Kaiserreich (video game mod), which has only 7 sources total, where only 3 are about Kaiserreich itself. The article on Kaiserreich isn't being nominated for deletion: what makes it's sourcing better? Jerryeit (talk) 17:07, 20 July 2026 (UTC)[reply]
I would say that the sources have more info extracted from them, making the article more expansive. Though, as I stated earlier, Create has been covered enough that it should pass WP:GNG just on the reliable sources alone IMO, but you may still try to extract more info from them. Dabmasterars [RU/COM] (talk/contribs) 21:38, 20 July 2026 (UTC)[reply]
Merge to Minecraft modding. While it has several mentions in reliable sources, it still suffers from a lack of significant coverage. It's absolutely worth a mention, but it could use more coverage to fully ensure a complete article. ᴢxᴄᴠʙɴᴍ (ᴛ) 21:38, 19 July 2026 (UTC)[reply]
Delete. I agree that it contains LLM (though not 100% LLM as shown by human errors in grammar and punctuation): there are bad source/text mismatches as well as textual signs. Even if reliable sourcing could be found, this needs to be torn out and done over. M kuhner (talk) 03:53, 17 July 2026 (UTC)[reply]
Delete - AI slop could be fixed during the editing process, but even so this musician and his material are only visible in typical self-upload services and promotional sites that reprint his own press releases. No reliable media coverage and therefore not notable. Good luck to the kid as he gets started and learns more about how to talk to girls. ---DOOMSDAYER520 (TALK|CONTRIBS) 13:28, 17 July 2026 (UTC)[reply]
Disclosure: I have a connection with NupekoTV and have edited this article. I have substantially rewritten the article, added additional independent sources and revised the content to follow a neutral encyclopedic style. The article now includes coverage from independent publications including The Informant247, Sahara Reporters and CIIJ. I welcome independent review. Yabagilaf (talk) 01:37, 21 July 2026 (UTC)[reply]
Delete Training and logistics is an office job, this is a secondary position in that particular branch, nothing to confer notability. Civil servant doing their job. Oaktree b (talk) 14:32, 15 July 2026 (UTC)[reply]
Delete. Draftify. I have found a total of four reviews of her books: 1, 2, 3 and Choice: Current Reviews for Academic Libraries; Jul2024, Vol. 61 Issue 11, p1151-1151, 1/4p. 3 reviews of 1 book and 1 review of 1 book isn't sufficient for WP:NAUTHOR. HRShami (talk) 10:59, 15 July 2026 (UTC)[reply]
Based on the reviews David Eppstein has found I have changed my vote to Draftify. While she is clearly notable, there are large swathes of unreferenced text in the article and the OP has also raised concerns about AI-generated content in the article. HRShami (talk) 05:58, 17 July 2026 (UTC)[reply]
Keep. Beyond the reviews listed above by HRShami, I found three more, two of another book: . I think this is enough for WP:AUTHOR, and provides plenty of in-depth independent sourcing about her work for WP:GNG. —David Eppstein (talk) 23:02, 15 July 2026 (UTC)[reply]
Keep as per the reviews of her works in academic reliable sources identified in this discussion that together show a pass of WP:NAUTHOR so that deletion is unnecessary in my view, Atlantic306 (talk) 23:47, 15 July 2026 (UTC)[reply]
Delete. Salvaging the content is not possible either given the information present and how it's written. Agreeing with nom and their assessment. ABehjat (talk) 22:58, 14 July 2026 (UTC)[reply]
Hold up. This was published by Lane Raspberry, whom I rather doubt was using AI. It appears to have been made in an editathon. Why do you think it is AI? Was it introduced after Lane published it? If so, surely this can be salvaged by just reverting to an earlier version. Or perhaps Lane has insight into its creation? CaptainEekEdits Ho Cap'n!⚓23:15, 14 July 2026 (UTC)[reply]
For me, the section labelled `Neurodiversity-lite` sounded a lot like an AI article (basically like an essay transformed to a Wiki article template/format). ABehjat (talk) 23:25, 14 July 2026 (UTC)[reply]
Sounds good - I wasn't aware of the context, and was reviewing the most recent set of changes and it was a lot of back and forth between two users so I thought it was spam.
(If we restore to original created article) - Rename to "Neurodiversity in work environments" or "Workplace accommodations for neurodiverse persons" - title(s) better suiting the sources, since the sources don't mention "labor rights"
@Avatar317: I do not have a rename suggestion. There are Wikipedia articles or redirects for "labor rights by (country)", and I chose this title based on that precedent, which is an adequate fit but there could be something better. Bluerasberry (talk)13:49, 15 July 2026 (UTC)[reply]
Comment: I oppose a deletion, but whether to merge or move is up to you all. Everyone knows Blueraspberry, who ran for the WMF board, and by reputation would never create an AI article. Bearian (talk) 04:19, 21 July 2026 (UTC)[reply]
Keep, I agree with the above, frankly. Lane is certainly far from someone to use AI in their writing. The article also definitely seems noteworthy here, and I would oppose a merge as above. The topic of neurodiversity is far broader than just autism. EggRoll97(talk) 04:54, 22 July 2026 (UTC)[reply]
Article is a direct translation of the French Wikipedia article and appears to check out. What's the issue aside from the article creator? BeanieFan11 (talk) 21:08, 14 July 2026 (UTC)[reply]
Keep? The article appears to be a direct translation of the corresponding French Wikipedia entry and appears to check out. Yes the author of the article was blocked for LLM and other issues but it doesn't seem like those issues affect this specific article? BeanieFan11 (talk) 17:20, 20 July 2026 (UTC)[reply]
Relisted to generate a more thorough discussion and clearer consensus. Relisting comment: Any responses to Beanie's inquiry? Please add new comments below this notice. Thanks, StarMississippi03:34, 21 July 2026 (UTC)[reply]
@Ighs It actually explicitly meets both of those. Can you explain where you're drawing the notion from that it doesn't? It also satisifies secondary source requirements and breadth of coverage. The only backing you've provided for your claim is that it seems trivial, which isn't really an argument at all.
I will say (as someone who decidedly isn't one) that I think this is part of the reason why conservatives believe this site is biased against them. They have just as much history behind their movement as liberals and leftists do, and it's equally important that we explore it.
On WP:NPOL, Call has never held office, or been a member of a legislative body at that level. Coverage is mostly trivial (hired, elected, re-elected, election ongoing), but you could make the case for embezzlement articles meeting notability. In general (excluding that), coverage is trivial or not independent of the subject. Ighs (talk) 13:11, 12 July 2026 (UTC)[reply]
It seems like half the time Wikipedia editors expect every article subject to be an A-list celebrity; people necessarily exist in their context. He was the statewide chair of one of only two major parties in the United States. NPOL doesn't actually explicitly say "government" office, it just says statewide office. Nonetheless, one line down from the one you referenced would indicate that he's a [major] local political [figure] who ha[s] received significant press coverage. Were you hoping for an expose on the layout of his garden? You only need 2 sources to establish notability. Even without mak[ing] the case for embezzlement meeting notability, that's still been well and over satisfied.
Coverage is mostly trivial (hired, elected ,re-elected, election ongoing) − Challenged by R Street, CBA-CLE, US News
coverage is...not independent of the subject - You misunderstand what that actually means, per WP:GNG: "Independent of the subject" excludes works produced by the article's subject or someone affiliated with it. For example, advertising, press releases, autobiographies, and the subject's website are not considered independent. Four of 15 sources are/were affiliated with Call in some manner or another. None of them, though, were used to establish notability.
Call does, however, satisfy the earlier rule on "significant coverage" − that addresses the topic directly and in detail, so that no original research is needed to extract the content. Significant coverage is more than a trivial mention, but it does not need to be the main topic of the source material.
Keep: Coverage seems substantial, as far as I can see, meeting WP:NPOL's "major local political figures who have received significant press coverage". Additionally, the embezzlement and other issues appear to support further. --Engineerchange (talk) 14:58, 12 July 2026 (UTC)[reply]
My impetus, my muse! I've missed our chats. @Magnolia677 what's your estimate of how many state or national committees you'd have to chair before you're no longer considered WP:RUNOFTHEMILL?
Keep On balance, I think the subject clears GNG. There are stories about his election as Colorado Republican chair and his disbarrment. I am not sure (politician) is the right title, however. --Enos733 (talk) 18:09, 12 July 2026 (UTC)[reply]
I'd initially had it as 'party chair', but it was moved soon after.
Keep, Individual who is the subject of the article in question received significant coverage relating to disbarment in 2021 stemming from embezzlement. (AP, CBS, etc.) Individiaul also got some not significant coverage relating to an alleged illegal vote in 2011. In 2013, Individual got significant coverage from KUNC about their election to state party leadership. Also there is significant coverage from other sources online, such as this one.--RightCowLeftCoast (Moo) 21:55, 13 July 2026 (UTC)[reply]
Hey would you mind posting those sources in a Refideas box
({Refideas|[Url here (space in between) any name you wanna give it, at least a character here]})
Relisted to generate a more thorough discussion and clearer consensus. Relisting comment: Trending towards keep, but relisting to be sure - there is no real discussion about the depth and the quality of available sources. Please add new comments below this notice. Thanks, Sandstein 19:47, 20 July 2026 (UTC)[reply]
To be clear we've got 29 sources on this article now and at least 20 of them are from reliable regional or national sources discussing him directly in relation to his embezzlement case or the state GOP. I can't think of any reason why this wouldn't be considered a Keep, even if we need to strip it down to solely to a consideration of WP:GNG, which if I'm not mistaken only calls for significant sustained coverage by 2-3 reliable national or regional sources, which we have well and truly surpassed by this point, and I still have at least a dozen sources I haven't worked into the text yet. Also just so we're clear I in no way used AI to write this other than google search finding synonyms for words, so even if it's listed as such don't let that be a factor in your decision.
Like guys what is this, this has kind of gotten absurd. I'd get it if we were looking at a barely developed article with like 5 sources that only focused on his embezzlement case or minor events in the state party; I've written almost 2,000 words people, the train has left the station. Unless this is a training exercise in analyzing base principles to the point of nonsense I cannot see any reason we still need to be here. Once we've finished up, anyway, I'd figure someone on the Delete side must be able to complete its review, having looked over and considered all of the available sources to make their determination in this discussion.
Delete Using POL to show notability is rather generous, appears to have held various office positions. Getting kicked out over the theft of funds isn't quite meeting criminal notability either, appears to have a rather routine career otherwise. Oaktree b (talk) 00:50, 21 July 2026 (UTC)[reply]
Every politician is just doing their job. Notability has nothing to do with whether we personally find something interesting or intriguing but whether reliable sources took note of the subject and bothered to write about it. Katzrockso (talk) 00:38, 22 July 2026 (UTC)[reply]
Doesn't look AI generated to me, please provide more evidence. I'll check the sources for the text accuracy later, and if it's accurate, I'll vote keep. For now, I'm neutral. Dabmasterars [RU/COM] (talk/contribs) 20:27, 10 July 2026 (UTC)[reply]
The article uses the rule-of-threes several times (WP:RO3 in bold):
"These algorithms... generate lifelike images, videos, or animations"
"Users can customize bodies, clothing, and sociodemographic traits"
"The use of generative AI in the adult industry began in the late 2010s, initially focusing on AI-generated art, music, and visual content."
"These platforms enable users to create or view AI-generated adult content appealing to different preferences through prompts and tags, customizing body type, facial features, and art styles"
Other odd phrasings (suspicious portions in bold):
"The case aligns with California's recent legislation—SB 926, SB 942, and SB 981—championed by Senators Aisha Wahab and Josh Becker"
"catering to audiences seeking customizable experiences"
"These bills aim to protect individuals from AI-generated explicit images by criminalizing non-consensual distribution... empowering victims to report and remove harmful content from platforms"
"AI porn sites, therefore, attract those seeking unique or niche experiences, sparking debates on creativity and the ethical boundaries of AI in adult media."
Since you refered to an section from me "Users can customize bodies, clothing, and sociodemographic traits", thats on page 9, starting at the section The Current State of AI Porn, notably at "users at customize". I added the page information to the source. The Other Karma (talk) 12:03, 11 July 2026 (UTC)[reply]
I'm not Elon Musk, so I can't afford to get the necessary subscription to access the cited article, but the description is itself full of signs of LLM-generated content (suspicious portions in bold):
"—or AI porn—"
"Following systematic open coding, categorization, and inter-rater validation"
Keep. Unless stronger evidence of AI writing is presented. I'm not seeing any smoking guns. The diffs show gradual writing and addition of sources, without markdown, hallucinated refs, or other AI shibboleths. Rule of three and fronted adverbial phrases weren't invented by AI. Fences&Windows11:52, 13 July 2026 (UTC)[reply]
Why do you think that a human, working on a Wikipedia article (not a promotional piece), would use the rule-of-threes so extensively? Furthermore, I find that most of the suspected LLM-generated text was added by a single editor, Hypeabase, who made lots of incremental but still reasonably-large edits in late 2024. GrinningIodize (articles without a Wikidata item) (talk) 16:59, 13 July 2026 (UTC)[reply]
Keep. This is a classic case of WP:AFDNOTCLEANUP. I do get the point about "close paraphrasing" and dry, algorithmic phrasing but without evidence of LLM, it's a case rewriting sentences on the talk page. The topic is receiving extensive global coverage across legal, ethical, and technical secondary sources and it feels like a high-traffic encyclopedic topic which should be deleted because of a few sentences using the rule-of-three. Needs cleaning and rewriting but strongly in favour of keepMissBVCC (talk) 15:02, 13 July 2026 (UTC)[reply]
Delete as likely failing NOLLM. In addition to all of the AISIGNS already pointed out by GrinningIodize, there are multiple large edits with unreasonably short time in between them early in the article history, at least one of them has a hallucinated reference. ‑‑gurkubondinn16:46, 14 July 2026 (UTC)[reply]
KEEP - AFD is not cleanup and there is no clear signs of LLM usage - all I see here is poor phrasing. The main thing you're saying here is usage of the Rule of 3, but that is in fact a standard writing technique - LLMs using the Rule of 3 doesn't make everyone using the Rule of 3 an LLM. Jishara (talk) 14:50, 18 July 2026 (UTC)[reply]
Keep -- The initial versions were almost certainly AI-generated. The prose that originates in the initial 2024 versions indeed shows the exact signs you would expect from 2024 AI text, but if you refuse to believe the large amounts of research about AI-generated text, this is the article creator's userpage, in its entirety: Hi, Bobby here. I am an Artificial Intelligence enthusiast. I like what's going on with AI there in the world. I like creating stuff with AI. That being said, the article has evolved significantly from there, and the topic is notable; it's just those initial kernels that need a rewrite. Gnomingstuff (talk) 00:52, 19 July 2026 (UTC)[reply]
Comment: I am painstakenly going through every source (you won't believe how annoying it is to bypass the paywalls) and removing info that wasn't supported by it. Already done about 40%. The moment I will finish it all, I'll vote keep (and I will vote keep, since there are enough reliable sources that talk about the subject in detail). Dabmasterars [RU/COM] (talk/contribs) 10:35, 20 July 2026 (UTC)[reply]
@Dabmasterars, it looks like you're adding back text that is contaminated with LLM-generated content, with slight modifications from the pre-stubification version. If you must return the information conveyed in the LLM-contaminated revisions, would you please rephrase it in your own wording?
For evidence of LLM contamination in the latest revision as of writing, here are some suspect phrases that were reintroduced post-stubification (suspicious portions in bold):
"SD's public release led to dedicated communities exploring both artistic and explicit content, sparking ethical debates over open-access AI and its use in adult media"
"While popular among niche audiences, the influencers have prompted discussions about their authenticity and consent, blurring the line between human and AI-generated content."
"The case aligns with California's recent legislation—SB 926, SB 942, and SB 981—championed by Senators Aisha Wahab and Josh Becker and signed by Governor Gavin Newsom."
"enabling tailored sexual and non‑sexual interactions"
@Dabmasterars Furthermore, one of your edits to that article mentions removing Oxford commas, which would seem to go against the advice offered by WP:AISIGNS:
Please do not merely treat these signs as the problems to be fixed; that could just make detection harder.
Keep The subject passes WP:GNG. If there is an issue that the text isn't written natural, we can remove any WP:NOR and puffery. Per WP:LLM"the use of LLMs to generate or rewrite article content is prohibited". However, there are some exceptions, "editors are permitted to use LLMs to suggest corrections to their own writing, and to incorporate them after human review." It is very specific that it only covers spelling, punctuation, capitalisation, and grammar. See WP:LLMRESP. Issac I Navarro (talk) 23:20, 20 July 2026 (UTC)[reply]
While the topic itself appears notable, the article seems to be heavily written using AI. First drafted by a user with <50 edits at User:See U in world/sandbox: the first version of the page included no citations at all and displayed several other signs of AI writing including groups of three, etc.. More concerningly is that most of the prose written in this first version still makes up the majority of the current article's prose. Requesting draftifying the article until a through clean-up can be carried out. Johnson52406:54, 8 July 2026 (UTC)[reply]
Oppose. Issue resolved. The article has been thoroughly cleaned up, AI-like prose has been removed, sourcing has been improved, and the content has been rewritten in an encyclopedic style. See U in world (talk) 07:29, 8 July 2026 (UTC) Blocked sock. ‑‑gurkubondinn17:45, 21 July 2026 (UTC)[reply]
Delete as failing WP:NOLLM, the user seems to admit to as much in recent edit summaries. The page history of the article and sandbox both show rapid and large edits. ‑‑gurkubondinn11:42, 8 July 2026 (UTC)[reply]
Weak Oppose per See U in World, issue seems to be resolved and cleaned up, so draftification likely would do nothing to change article quality or information. R. M. Holda - (talk) 18:08, 8 July 2026 (UTC)[reply]
Comment If the initial text was from an LLM, much of the article is still from an LLM. The groups of three have been removed but the remaining text hasn't. Take the first paragraph in background for example - it's had the first line removed due to including a group of three but the rest remains the same.
The original article creator was the one to both add that paragraph and remove the first line with the note "seems like LLM Generated". I'd assume they would know. Salmon Of Ignorance (talk) 08:42, 9 July 2026 (UTC)[reply]
Keep—The article currently reads to me like... well, like a Wikipedia article. If there were tell-tale signs of LLM-generated text before, they aren't evident as of now. Kurtis(talk)09:19, 12 July 2026 (UTC)[reply]
The problem I have with this line of reasoning is the fact that many people who join Wikipedia and become active contributors are often already experienced writers, and some may have an authorial voice that is strikingly similar to that of generative AI. In fact, just a couple days ago, I made a post to my Reddit profile in which I directly linked numerous examples of my writing, some of which go all the way back to 2016. I did this because I had multiple different people accusing me of using AI for my comments on the basis of their formatting (i.e. the fact that I use em dashes). I really don't want us to get carried away with accusing people of using LLMs for editing on the basis of their writing style; if they wrote the article from scratch, having someone drag them over the coals for "AI editing" could make them feel alienated. I dislike that idea even more than the notion of us occasionally having AI-generated text getting past the radar. Kurtis(talk)14:52, 16 July 2026 (UTC)[reply]
Oppose The article looks way better now as an encyclopedic entry. There is no need to transfer this to draft article. Abishe (talk) 16:13, 12 July 2026 (UTC)[reply]
But have they been fixed? Since the AfD was opened, the article has mostly been edited by the user that added the LLM-generated text who has been removing the {{AI-generated}} tag with vague explanation and added sections with AIATTR. I'm all for AI-generated articles being fixed and kept, but I'm not sure that it has happened here. ‑‑gurkubondinn13:35, 15 July 2026 (UTC)[reply]
Comment (as nom): I've removed and requested that the speedy tag not be added back so that this discussion can run its course since its been ongoing for about half a month. Johnson52413:44, 21 July 2026 (UTC)[reply]
This article appears to mostly have been generated (potentially by an LLM) from two academic papers, and conflates some of the content of the papers with the specification itself. I removed an example of this in an attempt to improve the article, but I now think the required changes are too significant and it should be deleted.
The structure of the article seems to be inferred from the content of the academic papers, rather than being helpful for the reader. Greg Tyler(t•c)12:07, 3 July 2026 (UTC)[reply]
Keep. I've stubified the article to remove the conflation of conventional commits with the methodology used in the papers. ~ A412talk!14:36, 3 July 2026 (UTC)[reply]
Comment: the creator's edit summary outright acknowledges this was created using Gemini AI and manual editing, so an LLM was definitely involved at some point (even if the remnants of its output have been at least reduced in the stubification). No opinion. WCQuidditch☎✎15:07, 3 July 2026 (UTC)[reply]
Relisted to generate a more thorough discussion and clearer consensus. Relisting comment: So what do we have left and how does it relate to GNG? Please add new comments below this notice. Thanks, SpartazHumbug!07:47, 19 July 2026 (UTC)[reply]
User:Alaexis/AI_Source_Verification (source) - user script to check citations, either one-by-one or across an entire page. Can use a variety of models on the backend, including open-weight models.
Wikipedia Tree Browser (source) - jina-embeddings-v4 + gpt-oss-20b clustering for hierarchical visual browsing of Wikipedia topics (219 commits, most actively developed find)
CommonsVibe - Pinterest-style feed explorer for Wikimedia Commons categories
