Cloudflare fetched the original article and streamed it through HTMLRewriter to apply an entirely new visual system without rebuilding the source page.
// request.cf · coarse context
A page that knows where it met you.
Only coarse request metadata is shown. This demo does not display or persist visitor IP addresses.
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.
Delete. Should note that the article creator's only edits that aren't this or link suggestions are mostly similar AI-generated articles about disasters in the China region, including one about the Ghost Glacier collapse in Nepal, in their sandbox. AdoTang (talk) 22:59, 23 September 2026 (UTC)reply
This article is very certain to be AI-generated due to multiple instances of failed verification. For example, the source for the sentences Since its inception, the PEARS procedure has been performed in several hospitals across England, demonstrating its potential as a viable treatment option. The procedure has undergone health technology appraisal by the National Institute for Health and Care Excellence (NICE), further validating its efficacy and safety doesn't say anything about Belgium (as it did in the version I PRODed), and the term "validate its efficacy and safety" is an exaggeration of its actual conclusion, which is that Evidence is adequate on the short-term safety and efficacy of personalised external aortic root support (PEARS) using mesh to prevent aortic root expansion and aortic dissection in people with Marfan syndrome. Evidence on long-term outcomes is limited in quantity and quality. For this claim (Clinical outcomes for PEARS have been favorable. In over 200 patient-years of follow up, there have been no untoward events related to the aortic wall or valve after PEARS. When compared with data for either the Bentall procedure or valve-sparing aortic root replacement, outcomes so far are very favorable for external support.), the source mentions 200 patient-years of followup but does not say anything about no "untoward events," nor does it mention the aortic wall. It does not make the claim that "outcomes so far are very favorable for external support" compared to the Bentall procedure (and which is a bit of a nonsense string of words anyway). It does not use the term "favorable" or "favourable" for outcomes. A non-existent DOI link in an earlier version ( is also a flag for LLM use. A PROD was contested so now AfD is the next step. While I believe this is a notable topic worthy of a human-written article, I have no interest in spending volunteer time checking every fact in this bot output and I believe WP:TNT is the appropriate response per WP:NOLLM. Dclemens1971 (talk) 16:23, 23 September 2026 (UTC)reply
"...personalised support has now been implanted in 46 patients in three English hospitals and one Belgian hospital..."
This in any case is an old link and it has been conducted in many more hospitals in the interim. More than happy for links to be updated, in fact there is a lot more research on this topic in the time since it was published. But deleting an article of relevance because the source for one sentence is actually in the next sentence does not seem a good approach. How much of wikipedia is being deleted because of this? Tororro (talk) 16:37, 23 September 2026 (UTC)reply
That kind of blending of statements and sources creates an unverifiable stew of claims, and finding two whole paragraphs not supported by the citations used on them is a good reason to TNT the whole thing and start over rather than piecemeal fixing them. You've removed the items I've identified, but what I've identified is just a small sample of the article's likely problems. Dclemens1971 (talk) 16:50, 23 September 2026 (UTC)reply
Comment Also, this is a medical topic and thus WP:MEDRS applies. Statements that are not supported by the cited references, even if those sources do meet MEDRS, is a very big problem in an article such as this. Remember, Wikipedia is where the LLMs look things up. WeirdNAnnoyed (talk) 22:37, 23 September 2026 (UTC)reply
This article is the second article connected to Antonino Cardillo which is written in a very obvious AI-style. I strongly suspect that the article is authored by somebody connected to the subject. There are two architectural magazines linked here giving sigcov, I am not an expert in that field but I suspect they may be reliable. Ordinarily I would stubbify, however, I am loath to leave an article in mainspace that may have been professionally commissioned without a declaration.
This article was apparently written by AI, in violation of WP:NEWLLM. I've made attempts to do clean up and verify sources but more than one has been a dead link or possibly AI generated. I tagged them for reference, but stopped after the lead. Other sources are not reliable, for instance a blog or self published. The subject may be a notable poet, but the article should not exist in the mainspace in its current state. Se7enNationArmy2024 (talk) 18:22, 22 September 2026 (UTC)reply
Fails WP:NSPORT and WP:GNG. Lower league player mainly covered in club websites, local match reports, etc. No significant coverage in published reliable independent secondary sources. Google search gives mainly local football reporting, social media, etc Paul W (talk) 13:00, 21 September 2026 (UTC)reply
Fails WP:NSPORT. Journeyman lower league striker mainly covered in club websites and in routine local match reports. No significant coverage (in-depth, focused) in published reliable independent secondary sources with reputations for accuracy and fact-checking. Google search confirms mainly local football reporting, Facebook updates and other social media mentions Paul W (talk) 12:34, 21 September 2026 (UTC)reply
My G15 request was processed, but the article was just recreated instead by the same user. My opinion is still to delete and oppose a merge, for the same reason. ‑‑gurkubondinn21:39, 23 September 2026 (UTC)reply
Delete and oppose merge per gukubondinn, and the fact the the article cites no sources means, ipso facto, there is no verifiable material worthy of merging into the main article anyways. Finally, editorially speaking, anything in this article would be overly detailed and WP:UNDUE to be included in the main article.
Obvious AI creation. Strange section header capitalisation, information arranged into bullet lists. The most telling sign is the sourcing format, wildly atypical from the author's easily identifiable messy usual style .
Edward Mike005 has long been a productive editor and a clear net positive. But they will be taken to WP:ANI if a second episode of AI usage is repeated. It is a block-worthy offense per WP:LLM. We are being swarmed by AI slop in the website and don't need editors not willing to put in serious effort. That is unless they can refute the allegations at this AfD. As a disclaimer, I haven't noticed the use of AI in any of their previous articles. SuperΨDro22:24, 21 September 2026 (UTC)reply
I appreciate your concern, but I would like to mention that it is not "AI slop". Not at all.
While I use AI to write drafts, the final form of the article is re-edited by me, checked by me, and published by me. No bots or anything else.
The citation style is mine. I don't know that well to add citations that are not from the websites. Regarding this article specifically, the citations are added via the visual editor. The informations from the book of Iudean I mentioned.
I appreciate your concern, but I would like to ask you not to call me 3 hours work (that is the time it took me to write the article, if I hadn't use AI it would probably have taken 3 days) "AI slop". I use AI to write a draft. But the wikilinks, sources, and corrections - because, trust me, AI gets a lot of things wrong - are done by me. I do this to save time, and, I repeat myself - I check all the information to make sure that it doesn't become "slop". I use AI to help me write the articles, but the articles are not written by AI.
Update: I was wrong, they have made use of AI before . May this AfD serve as a warning to the editor. WP:LLM states clearly the use of AI in Wikipedia is allowed only for basic copyediting and translations, both requiring subsequent human reviewing. SuperΨDro22:30, 21 September 2026 (UTC)reply
Draftify: The editor made an honest attempt at an article, I'd allow this to go to draft space to be corrected and written properly. AFC can be a great tool for newer editors. I think the intention is there, perhaps the execution leaves to be desired, but I'd not "bite the newbie" over this. To be blunt, LLM is a NO GO here, but I hope the editor can learn from their mistakes and will continue to contribute. All are welcome and we need good, hardworking editors in Wiki. Oaktree b (talk) 18:50, 22 September 2026 (UTC)reply
Agreed. I apologise if my initial tone may have been excessively harsh. The editor really is a net positive. But one of the few things I think could ruin the project is the excessive use of AI. Let's keep things here authentic. SuperΨDro22:08, 22 September 2026 (UTC)reply
like I said: while AI was used for the draft, the final article is corrected BY ME. I check every sentence, every word, to be factually correct, to be written correctly.
The internal links and wikizations are also made by me. And the sources as well.
I didn't told the AI to write me an article then copy paste here. I used what it wrote as a draft that I expanded, corrected, fixed.
Fails WP:NSPORT criteria - insufficient significant coverage in published reliable independent secondary sources with reputations for accuracy and fact-checking (so excludes local football sites, club reports, databases, routine listings, stats, etc) Paul W (talk) 09:50, 18 September 2026 (UTC)reply
Delete - all coverage is either not significant, not independent or not reliable, hence does not meet GNG. He has not played above step 4 so also seems unlikely that will change. Microwave Anarchist (talk) 22:15, 19 September 2026 (UTC)reply
I believe this article features unreliable sources and claims that have not been confirmed, as well as being written similarly to a LLM, which violates article policy Krazonyx (talk) 11:01, 21 September 2026 (UTC)reply
Keep. This article is adequately sourced, as reflected on the article's talk page, which has been reviewed and updated since the article's move to mainspace. It was reviewed and accepted from draft via the WP:AFC process on 21 September 2026, the same day this AfD was opened, indicating an independent reviewer already found it met the relevant content standards. An unsubstantiated claim about AI-assisted writing style is not a valid deletion rationale; deletion discussions are meant to focus on notability and verifiability per WP:GNG and WP:V, not on stylistic speculation. If specific sources are believed to be unreliable, I welcome them being identified so they can be addressed directly.Saji•talk11:56, 21 September 2026 (UTC)reply
This nomination is weird because it doesn't explain which sources would be unreliable and which claims would be unconfirmed. Maybe it's not a great idea to try AfD after only 20 edits.
At the same time, this reply is also weird. What is reflected on the talk page? There's nothing much there about the sources. Likewise, what I told you about AI is very much substantiated - several of the sources were only linked to the root URLs of those websites or not linked at all, which is something the LLMs do, but it's inconceivable that a real editor would do it, so the use of AI was pretty obvious. You also failed to simply deny that you used AI, and have continued to handwave around that issue, which is a bit cringe at this point. --Joy (talk) 13:17, 21 September 2026 (UTC)reply
The nominator suggests three main reasons for deletion: unreliable sources, unverified claims, and LLM-like writing, none of which are reasons for outright deletion. This seems like a matter for the talk page rather than WP:AfD. guninvalid (talk) 14:30, 21 September 2026 (UTC)reply
This has already been addressed in detail here , it relates to a single early (July 2026) incident, disclosed and resolved at the time, unrelated to this article. Saji•talk10:37, 22 September 2026 (UTC)reply
Keep I'm sorry, I just don't buy that it's AI, at least based on the evidence we have. First off, automatic notability per ANYBIO, so that's not even a question. As for the sources, the one without a link was a journal which is frequently not accessed digitally, the one that went to a homepage appears to automatically do that (like certain other sources) when put through the default Wikipedia citation tool, and the way the sources were written at creation doesn't read as AI (i mean, the cite web, link - before the site name and a link instead of a source name screams the default citation tool, it does it like that all the time in my experience, especially for non-English sources). Jishara (please ping upon response) (talk) 23:11, 21 September 2026 (UTC)reply
And you think an editor who is reviewing AI output wouldn't notice that half their links are broken? Come on now. Which isn't to say that this can't be easily cleaned up, but we shouldn't try to excuse this shoddy work and pretend all is well, rather we should simply fix it, learn from it and know better for next time. --Joy (talk) 11:28, 22 September 2026 (UTC)reply
Comment: the section heading "Academic division of the University of Zadar" and its use in that paragraph is puzzling and suggests that the editor may have mistranslated a source, as "Academic division" seems to be the equivalent of faculty (UK) / school (US) / department, and does not stand alone. This casts doubt on whether the text has been written by a human editor who understands their sources. PamD07:40, 22 September 2026 (UTC)reply
This article is translated and adapted from the Hungarian Wikipedia article; no such section heading exists in either this article's Hungarian source or the Croatian Wikipedia article on the subject. That section heading isn't translated from any source. I wrote it myself as an organizational heading. Happy to change it to "Department" or "Faculty" if that's clearer or more standard terminology.Saji•talk08:08, 22 September 2026 (UTC)reply
@Sajjadmrt The fact that you used it like that suggests that you are not sufficiently familiar with English to be writing articles in en.wiki or translating articles into English. PamD13:34, 22 September 2026 (UTC)reply
@Sajjadmrt You also stole another editor's, or other editors', work and passed it off as your own when you "wrote" this article on 19 September without any acknowledgement of it being "translated and adapted". PamD13:36, 22 September 2026 (UTC)reply
This tone is not conducive to a collegial discussion. You have made personal attacks against me; per WP:CIVIL and WP:NPA, comments should address content, not the editor's competence or character. I'd note that my work has received support from numerous editors and reviewers throughout this process. On the attribution point: I added the translation attribution note to the talk page as soon as the section heading question raised the underlying concern this was proactive, not concealment. The article's content is accurate to its source regardless of when the attribution note was formally added. Saji•talk14:32, 22 September 2026 (UTC)reply
It's proactive to own up to something three days after having done it?
And the question of attribution isn't an issue of mere accuracy, it's about a policy with legal considerations. We know it's hard to reconcile that with the modern-day AI logic, but you need to do it nevertheless.
Fundamentally, people will inevitably get annoyed when you fail to observe policies and guidelines, refuse to take responsibility, and then apparently keep wikilawyering in response. --Joy (talk) 14:03, 23 September 2026 (UTC)reply
Does not meet WP:GNG or WP:NACADEMIC. No significant coverage of the subject available in independent reliable secondary sources. The sources available are largely employer profiles, routine mentions, and papers co authored by the person. The article also does not demonstrate that any criterion of WP:NACADEMIC is met. Also, several citations currently used in the article are unrelated or do not substantiate the claims they follow. - The9ManTalk08:46, 21 September 2026 (UTC)reply
Potentially LLM written article that's mostly just an essay about a vague and broad topic instead of an article about a well established concept. "Immigrant entrepreneurship" can be anything. Cubnorth (talk) 23:33, 20 September 2026 (UTC)reply
Keep with reduction. Immigrant entrepreneurship is a widely accepted term and the subject of a substantial body of academic research. It is specifically studied and tracked by organisations including the OECD, which includes a dedicated section of its International Migration Outlook 2 to the subject, and is the focus of a specific chapter in Elsevier's Handbook of the Economics of International Migration, among many other major academic publications. This is precisely the type of significant subject that Wikipedia should include. Its prominence can be quickly verified through searches of Google and JSTOR.
Wikipedia contains many obscure subjects that may legitimately warrant articles, but this is an established and important area of study that deserves inclusion because of it encompases such a broad range. The sources I have provided are all strong academic publications devoted again specifically to the subject. Even the source provided by ladmc, the German Historical Institute's Immigrant Entrepreneurship project covering immigrant entrepreneurship in American history broadly, reinforces the importance and established use of the term.
That said, although I am strongly opposed to LLM generated content, I recognise that my writing style may resonate with this issue. My writing also tends toward an essay style, which is something I have done professionally since long before AI existed and is a legitimate complaint by other editors. I am trying to work this out of my writing in general. I believe this may be partly because I develop highly polished articles. I made nearly 300 edits on Wikipedia developing this article, (not what you find in LLM generated content) and probably an equal amount of the work was done outside Wikipedia because I find it considerably easier to structure, move material around that way.
My proposed solution is the normal one for a notable subject where the article faces such concerns, that is to substantially reduce it. Over the next few days, during this discussion, I intend to shorten the prose considerably, probably by about half, while retaining as much of the existing structure and sourcing as possible.
One problem on Wikipedia is that articles can grow through incremental additions until they become sprawling and unorganised. I would like to leave this article with well developed structure but limited prose that makes it conducive to future contributions by other editors, while retaining the substantial academic sourcing that provides a foundation for future expansion. This is an approach I plan on taking in all of my future writing, limited prose with strong structure.
I want to elaborate a little more as the nominator. There is no one unique thing that ties the information in this article together except for 'immigrants can be entrepreneurs'. This means that in my opinion, that's all the article is really saying. If the observations that followed were conclusive and uniform, then I'd understand a bit more, but as it stands almost all of them boils down to 'it varies', which is obvious. I respect that this is an area of interest for you, but as it stands I believe this topic is far too broad to be a useful article. Cubnorth (talk) 13:24, 22 September 2026 (UTC)reply
I understand the concern that the subject is broad, but I think that breadth is also what makes it valuable as an encyclopedia topic. Wikipedia benefits when it covers important central concepts, not just narrowly defined subjects. The study of immigrant entrepreneurship has considerable importance within sociology and social economics, which is why there is such substantial academic literature devoted to it.
The difficulty for me has been presenting the body of material without turning the article into an essay or adding my own synthesis. I could write a more cohesive argument connecting the research, but that would not be appropriate for Wikipedia. The article needs to be built from what the sources actually say, which can make it difficult to present the material cohesively.
At the same time, I am addressing the perception of LLM-assisted writing in the typical manner, by substantially reducing the prose and making the article more direct and source-focused. This often works against your concern, since reducing the prose can also make it harder to explain the connections among the research. I will post a rewritten version in the next few hours.
I understand where you are coming from, but I think this is a case where creative editing and restructuring can solve the problems. It would be a shame to lose an important area of study that belongs in an encyclopedia when the underlying subject and sourcing are very strong.JondeJon1 (talk) 08:05, 24 September 2026 (UTC)reply
I previously nominated this article for speedy deletion on WP:G15 basis, which failed. Why I believe this article was written by an LLM:
Style: "nomenclature corresponds exactly to its mathematical position inside the historical currency matrix", "proved structurally incapable", "clean structure", etc.
Formatting: lists (where humans usually choose subheadings) and bolds (e.g., Heaton Mint and Hamburg Mint).
Sources that have no relation to the paragraph they are put next to. For example, has nothing on "issued by local gambling houses (rong huai)" or on "The traditional hand-stamped manufacturing of bullet money (pod duang) proved structurally incapable of keeping pace." Vgbypw (talk) 15:05, 20 September 2026 (UTC)reply
Delete: I don't find any sources that talk about this, the half-pai either. I only get hits on the phrase... Numismata and wiktionary are not RS. I don't see notability here. Oaktree b (talk) 15:47, 20 September 2026 (UTC)reply
Keep. It seems the concerns raised are regarding the content added on 21 May. In that case, at worst it should suffice to revert to the prior version from January without need for deleting the article, which is part of a series, the rest of which don't appear to have been flagged. --Paul_012 (talk) 15:50, 20 September 2026 (UTC)reply
If notability issues are to be discussed, then this should be a mass merger nomination covering all the pre-decimal Thai coin articles. --Paul_012 (talk) 15:52, 20 September 2026 (UTC)reply
Delete: I decided to look into the sourcing since Oaktree b brought it up. The article's sources are a journal article (is reliable), Numista (a not reliable WP:USERGENERATED website), Wiktionary (certainly not reliable), and Dema Coins, which seems to be a WP:SELFPUBLISHED website or blog by someone named Dema Yurii whose numismatic expertise I cannot find in any independent publications (although it does seem to provide in-depth coverage). I will note that much numismatic research is done by amateurs. I have looked for coverage in reliable sources (in English, I imagine there might be more in Thai), and have found only these:
Entries in The Standard Catalog of World Coins 1900-2000 and A Catalog of Modern World Coins 1850-1964 (I physically own these books)
Ignoring the LLM stuff, I am not sure if it has WP:SIGCOV. Coin articles for currencies in non-English speaking countries are extremely tricky to find sources for. I would like a definitive yes or no if descriptions like the one in the The Numismatist and The Standard Catalog of World Coins count towards establishing notability, as lot of coin and currency articles use them to establish notability for denominations of coins. I will note that the creator of the article User:Nicholas Mah has stated on the talk page that he did not use an LLM to write the article. I agree with Paul_012 that a wider discussion on Thai coinage articles needs to happen, especially on Nicholas Mah's excessive contributions for Thai currency topics (see the state of Thai baht right now).
No reliable independent sources show that the subject meets Wikipedia's notability guidelines. The article mainly relies on primary sources and social media, which are not enough to establish notability. Tyelksa (talk) 07:18, 17 September 2026 (UTC)reply
Delete; Keep; there are several sourcing issues (like noted with primary & user-generated sources) However, there are some reliable sources demonstrating notability. Though the article could use work, it seems to meet GNG. I changed my mind after seeing WP:NEWSORGNIGERIA and WP:SYO. WikipediholicRat (talk)14:07, 17 September 2026 (UTC)reply
Keep: Being him as the chairman of the Nigeria Diaspora Sports Council (NDSC) could establish notability plus independent reliable sources from , , , and are enough to establish notability. Thus pass WP:GNG and WP:SIGCOV~2026-49181-30 (talk) 21:46, 19 September 2026 (UTC)reply
Delete - while the BBC Sports is a reliable source, the article is about the team of which he is "assistant coach and head scout". Neither a head scout nor a chairman of a sports association for diaspora clubs are notable per se. The remaining sources are blogs and local news outlets that might be less reliable. Therefore, there is a lack of significant coverage. It might just be too soon in his career for notability. Bearian (talk) 15:07, 22 September 2026 (UTC)reply
I do not see a pass of WP:GNG for this local Indian mayor. The only coverage that exists about him is WP:ROUTINE and WP:MILL coverage of his election as mayor, the same coverage any mayor would get. Per WP:NPOL, local political officials are not inherently notable. Beyond that, this article deserves WP:TNT for vague LLM-generated language (His election marked a notable development, with some commentators highlighting its significance within local and regional politics, reflect diverse representation in civic leadership) Dclemens1971 (talk) 02:52, 20 September 2026 (UTC)reply
keep
The article is of an real mayor of an city in State of Maharashtra, the content looks generated by AI because it is generic so the article can be improved rather than deleted. —Preceding unsigned comment added by Eaglespirit (talk • contribs) 03:44, 20 September 2026 (UTC)reply
Likely LLM-generated article that isn't a "glossary" at all; it's a list of books on caste and authors who have written on the subject (plus a few keywords like in an actual glossary). Essentially edited by a single user since creation last year. Fails WP:NOLLM, WP:NLIST and WP:SYNTH. WeirdNAnnoyed (talk) 20:27, 19 September 2026 (UTC)reply
Fairly sure this is LLM-written and I don't think it's really worth attempting to clean up. Also fairly promo but not sure if that's just the LLM or some undisclosed COI involved there. Alpha3031 (t • c) 14:28, 19 September 2026 (UTC)reply
Comment: draftify might work if someone who knows about the brand can clean it up. There do seem to be a lot of references, though I never checked their relevance. If it is just a load of COI self-promotion, then delete. —Iadmc♫talk15:48, 19 September 2026 (UTC)reply
Delete I analyzed the media sources, and it is clear that an LLM was used to generate the article (many of the sources mention this in the link ?utm_source=chatgpt.com). I verified all the sources and identified several blogs, which I removed from the article, as well as a number of PR sources that cannot currently demonstrate notability.Latitudemix (talk) 11:48, 21 September 2026 (UTC)reply
The topic covers an aspect of generative AI in Dental and medical education which is growing; if there a is a similar page covering the same aspects, please suggest. Faraortho (talk) 21:40, 19 September 2026 (UTC)reply
The nomination appears to be based primarily on suspected LLM use rather than an examination of the subject's notability. The article can and should be evaluated against Wikipedia's notability requirements and the reliability of its sources.
Patrick Amofah meets the basic framework of WP:GNG/WP:BIO: there is significant coverage of him in multiple published secondary sources that are independent of the subject.
Examples include:
Citi Newsroom (2022), which profiles Amofah, his background and the development of HourHand Watch Company.
Citi Newsroom (2023), which provides substantial coverage of Amofah's work as a Ghanaian horologist and his servicing of a Richard Mille RM 011-03 McLaren timepiece.
Daily Graphic (2022), which profiles Amofah as CEO of HourHand Watch Co. and discusses his entrepreneurial career.
Graphic Online (2023), which independently reports on Amofah presenting a custom Warrior King timepiece to the Asantehene.
GQ South Africa (2023), which published a substantial profile/interview concerning Amofah's background, horology and Warrior King Watches.
YEN.com.gh (2023 and 2024), which published substantial profiles specifically concerning Amofah's career as a watchmaker and entrepreneur.
3Xtra (2023), which published a dedicated feature on Amofah and his work as a watchmaker.
Ghana News Agency (2025), which independently reported on Amofah presenting bespoke solid-gold timepieces to the Asantehene and Lady Julia.
These are not merely the subject's website, social-media accounts or press releases. They include independent Ghanaian news organisations and an international magazine, and several provide substantive coverage of Amofah himself.
I therefore believe the relevant question for this AfD is whether the subject has received significant independent coverage, not whether an editor's citation formatting or writing style resembles LLM output.
If there are concerns about individual citations, those can be corrected or replaced without deleting the article. I have no objection to removing the `utm` parameter from any affected URL, correcting retrieval/publication dates, or replacing any citation that does not meet Wikipedia's sourcing standards.
The existence of a `utm=chatgpt.com` parameter is not, by itself, evidence that the article's content was generated by an LLM. The appropriate test remains whether the article complies with Wikipedia's content policies and whether the subject meets the applicable notability guideline.
On the evidence above, the subject has received substantial independent coverage over multiple years. I therefore request that the AfD be evaluated under WP:GNG/WP:BIO rather than inferred authorship from URL parameters or citation formatting. MadeInGhana (talk) —Preceding undated comment added 14:58, 18 September 2026 (UTC)reply
Delete. The article is LLM--there are multiple lines of evidence both technical and stylistic. My experience is that fixing such articles is markedly more difficult than writing a new one from scratch: the only thing that draftification does is create a temptation to make cosmetic alterations and republish. WP:TNT is the best approach to this type of material.
I am also far from convinced of notability. A one line "presented a watch to a famous person" is not substantive coverage, and piling up examples of it still does not make it substantive. Can someone point out three sources that actually talk about the person in some detail?
I am a bit cofused here. Today started with the user:Biruitorul commenting on my article that it uses LLM. I did not use LLM. I never use LLM. And if this is so easly being thrown on editors here as a stamp, it is a form to gag anyone unfairly. I do not want to start an argumet with user:Biruitorul. I just want to state clearly that there was no LLM used in what I personally wrote creating this article.
This is based on my own, long-time work. If it is not wanted here, alright, but if it is improvable, please guide me. I understand by looking at his activity that user:Biruitorul appreciates very short pages and basic research, but this is not why I joined Wikipedia long time ago. I already have gathered research on other activists and artivists, like Rober Obert, Cristian Mihai Dide, Alexandra Gulea, Pisica Patrata, and I would like to create more pages here. But I do not want to risk my work being randomly deleted or falsely labeled LLM written.
If I got carried away in editing, as it was implied, maybe someone can give me some guidance as to how to improve.
PS: in the short time I left this article unsupervised, mistakes started to pop up: someone added somehwere a date of birth of the person. Someone else added yet another date of birth in another place. At the moment when I am writing this comment, the contradictory dates of birth co-exist freely in the article I so carefuly created. What I want to add is that myself, I didn't add any date of birth for Madalina Rosca, not because I am a sloppy researcher, but beacuse there is't any clear and reliable information online from my point of view.
There is nothing in the article I initially wrote which does not have a reference. If my writing per se is the problem I am willing to learn, if anyone is willing to help or to actually comment other than labelling for deletion LLM or "promotional language".
So I agree, maybe I am getting too much into details, but I do not think it is fair to the very serious research I do, you can call it OCD maybe, to just say: delete. Let me know what to do better but please don't use horrid and false accusations like "LLM was used" or offences like "this is not salvagable" after so much work. Brutus65 (talk) 23:53, 17 September 2026 (UTC)reply
Draftify per above discussion; this might be a notable topic, but this is a very poorly written article. I taught legal writing for almost 2 decades to paralegal students, and see a lot of common mistakes. To be honest, it needs a lot of work, but the Guild of Copy Editors will refuse to work on any article that is tagged as having notability concerns. So the only alternative to deleting it is to send it back to the creator to work on it, with or without any help, for a few weeks or months. My unsolicited advice is "less is more". Don't expect me to help, as I'm too busy in real life. Bearian (talk) 11:14, 23 September 2026 (UTC)reply
AI-generated article on a journalist not shown to have met the WP:GNG. It seems like the most coverage he has received was when a politician told him to shut up in an interview. He's also won some awards, none of which I believe are "well-known and significant" enough to meet WP:NBIO. JTtheOG (talk) 21:39, 17 September 2026 (UTC)reply
I just declined the A7 as that does not apply IMO (courtesy @Scope creep:). While this likely did make use of LLMs, it also isn't a G15 and we could clean it up if sourcing was available, which brings me to the nomination. I'm having trouble finding sufficient sourcing to indicate he meets N:PROF or is notable as a doctor so bringing it here for discussion. StarMississippi17:38, 17 September 2026 (UTC)reply
Non-notable company; fails WP:NCORP. Sources are WP:CORPTRIV about capital raises (, , , , , , , , ) a press release (), an affiliated source (), and a trivial mention (). I don't see anything else qualifying in my BEFORE search.
Draftify or Keep. I agree with the nominator's assessment here. None of the sources individually are significant coverage, and the sources considered together do not constitute significant coverage. That said, I think this company is significant - from a brief search, I found this article from business wire (unsure if we consider it reliable for these matters) that says this company is the First European Quantum Computing Company Listed on a Major U.S. Exchange. Improvements need to be made, but I think that it is significant enough for inclusion! Redvelvetvanilaaaaaaaaa (talk) 15:44, 16 September 2026 (UTC)reply
Keep. While the company is in the initial stages, such is the case for most quantum computing companies that are not Google or IBM or Microsoft's quantum branches. Having a PhD in quantum technologies myself, and having attended quantum engineering conferences recently where the number of mentions of IQM quantum computing was very many (tens if not a hundred). Comparing to contemporaneous QC companies such as Quantinuum (founded 2021) and IonQ (founded 2015), IQM Quantum Computers was founded in 2018. The other companies have extensive articles already. So the spirit here should be to improve this IQM article instead of deleting it. It is indeed a pioneer of the European QC company ecosystem. To support my case, I put more secondary sources that satisfy WP:CORPDEPTH and are not authored by the CEO or co-founder of the company (some initial sources in the article were indeed press releases by IQM)
Also It is highly likely that there is some sort of Sockpuppetry Going on here as the TA that recreated the article overlapped with Many of the article Vlad edited on with the same edit summary of "Corrected the Mistake" as per: . BengalMCtalk09:47, 16 September 2026 (UTC)reply
Keep the NDTV profile in there is solid, as is the Himalaya Times piece and this Cricket Monthly piece. I'd recommend that we revert it to an older version, though - the current one has WP:AIBOLD and utilizes the * signs at multiple places, which is a pretty strong sign that it's significantly altered by AI - looks like it was during the edits on 21 January 2026, so revision to the version before that would work well Jishara (please ping upon response) (talk) 14:52, 16 September 2026 (UTC)reply
Comment. The issue isn't whether there are any sources mentioning Ghimire; there clearly are, Jishara. The relevant question is which sources constitute significant independent coverage for WP:GNG. The Cricket Monthly/ESPN feature appears to be the strongest source and is substantially about Ghimire, while several other results I found are primarily routine coverage of CAN events in which she is mentioned in her official capacity. I will pass this on to other editors' judgement. Nirvana869976 (talk) 20:02, 17 September 2026 (UTC)reply
Keep: Since the question raised in this AFD is WP:GNG, a 5 second google search gave me this on the first page which easily indicates she is notable enough and pass WP:GNG. I also agree with Jishara on AI . Khagendra (talk) 08:36, 17 September 2026 (UTC)reply
Comment. I don't want to influence you to change your opinion, Khagendra. However, I don't think the links you shared necessarily establish WP:GNG by themselves. Some appear to be routine coverage or mentions in connection with her official roles, rather than significant independent coverage of Ghimire herself.. But I will dig deeper onto what Jishara mentioned. And again, trivial mentions or routine company coverages don't qualify a person for WP:GNG. Nirvana869976 (talk) 20:01, 17 September 2026 (UTC)reply
Delete: Routine coverage of appointments and resignations, trivial mentions and other interview pieces do not help establish notability. Keep voters should try to find sources that are completely independent of the subject. New ground, old hurdles can be considered one such source, as it provides substantial independent analysis of the subject, but the subject is still far from meeting GNG. Jeraxmoira🐉 (talk) 14:26, 19 September 2026 (UTC)reply
One of a number of very lengthy, possibly AI generated, articles about the work of photographer Kansuke Yamamoto . This article is about a single photograph, and basically a description of what a single author says in her book Japan's Modern Divide: The Photographs of Hiroshi Hamaya and Kansuke Yamamoto. As such, Sleepy Sea shows little sign at all of meeting WP:GNG. Time for it to go. Sionk (talk) 17:56, 15 September 2026 (UTC)reply
Keep – The 2013 J. Paul Getty Museum catalogue discusses Sleepy Sea in detail. Amanda Maddox compares it with Magritte's The Reckless Sleeper and discusses the compass, pocket watch and toy gun marked "Japan". She relates the photograph to Breton, its title and postwar Japan. I think that satisfies WP:SIGCOV. WP:GNG does not set a fixed number of sources; its note on sources allows for cases where one source is credible, neutral and detailed enough for a comprehensive article. Maddox's essay appeared in an edited Getty Museum catalogue. The work is also documented in the 2001 Tokyo Station Gallery catalogue and in the 2017 Japanese Surrealist Photography materials. I don't think the nomination's reference to a single author, by itself, shows that Sleepy Sea fails WP:GNG when Maddox discusses the photograph in this much detail. Fontana2026 (talk) 22:35, 15 September 2026 (UTC)reply
Update: Since my comment above, I have added a 2013 CNN feature by Lauren Russell to the article. It discusses Yamamoto's political photocompositions and the 1951 security treaty, then describes Sleepy Sea as a 1953 work with a toy gun marked "Japan". It also says that the title does not say what it refers to. This is secondary coverage by a different author and publication. The nomination's point that the article essentially relies on a single author therefore no longer holds. Fontana2026 (talk) 19:14, 19 September 2026 (UTC)reply
Considering the Maddox essay is in an exhibition catalogue (with a vested interest in promoting the work) and the CNN source is a photo blog, with one small paragraph describing the photo, I stand by my nomination. A bloated article of repeated descriptions and context can be simply summed up by the three lines already in the Kansuke Yamamoto article. Sionk (talk) 22:06, 19 September 2026 (UTC)reply
Thanks for the reply. For WP:GNG, I don't think the fact that Maddox's essay is in a Getty Museum exhibition catalogue, by itself, establishes a specific vested interest in Sleepy Sea or makes the essay non-independent. Getty does not own Sleepy Sea. Maddox provides scholarly analysis of the photograph and its context. It is not promotional copy. Separately, on CNN, WP:NEWSORG notes that material from an otherwise reliable news organization can remain reliable in blog format. The CNN piece describes Sleepy Sea in a discussion of Yamamoto's political photocompositions and postwar politics. Fontana2026 (talk) 00:03, 20 September 2026 (UTC)reply
I have concerns that this topic, while potentially notable, is written in a way that merits WP:TNT. I checked the first five footnotes, and setting aside one which is dead and I can't open the archived page, none seems to mention this topic. The article may be an AI hallucination, or at the very least, much of the content here seems to fail WP:V. It doesn't help that there are parts, like the entire section on "Limitations", which are unreferenced. And also - Limitations is a section found in academic articles, but not usually Wikipedia ones; again implying major style problems. I will note that the fr wiki article is independent, and while shorter, does not seem to have such problems (i.e. the refs there seem to be on topic). PS. To illustrate the problem, consider the opening of the first section, which claims that "Coffee badging can be understood as a manifestation of status quo bias ", citing a reliable academic source . However, that source does not mention terms like "Coffee badging" or "status quo bias" at all. If I am correct, we may need to investigate more edits by the creator. Piotrus at Hanyang| reply here08:30, 15 September 2026 (UTC)reply
No evidence of notability for this film production company. All sources are passing mentions. This would probably be the best source, and even then it provides virtually no direct coverage. JTtheOG (talk) 23:19, 14 September 2026 (UTC)reply
I wasn’t aware of Wikipedia’s guidelines regarding the use of LLMs. Yes, I used an LLM to improve the articles. I assumed that AI could help me turn my initial drafts into articles more quickly. However, I understand the issue now and will gradually work through the listed articles to address the concerns and bring them in line with Wikipedia’s guidelines. Alternatively, I can recreate the affected articles with the issues resolved.
Draftify as the editor of the article has admitted to his use of AI to fix the errors and has reflected on it, the article can be re-written in Encyclopedic tone without the use of LLMs and more sources can be added to it. The subject itself seems significant enough to have coverage and a simple google search produces results. Farhanansari1994 (talk) 11:08, 17 September 2026 (UTC)reply
@Dreamyshade and @Farhanansari1994 Thank you very much for giving me the opportunity to salvage this article. The GPTZero result now indicates that the article is 100% human-written.
Please note that I have addressed the immediate LLM-related issues with the article, and I will continue working on it to further improve its quality and bring it in line with Wikipedia’s standards.
American motorsports promoter (BLP). Sources are mostly about his possibly more notable father, or about motorsports; there is nothing here that suggests sustained coverage of the subject in his own right in reliable sources, as required for WP:BASIC. The article has been tagged for promo, notability and verifiability since May.
Comment I think he might be notable, but the article is frankly horrendous and doesn't even say what racetracks he supposedly owns. Unsure as to what to do. ~2026-50236-34 (talk) 09:01, 17 September 2026 (UTC)reply
Comment The sources that may be about the company are paywalled, so it is hard to evaluate them. The reviewer declined the draft because of the product section, but I would say it is quite strong as the products are covered by the third-party outlets (unsure about reliability, however), so in principle, one can argue for the notability of the products based on the reviews. Draftification is fine, but the article is too old, so I am leaning toward deletion. Wikipedia.gptzero.me says that the page is human-written with 100% probability. Kelob2678 (talk) 09:27, 11 September 2026 (UTC)reply
I was able to reach a few of the product article and there was some substantial coverage. What I can't verify are how independent the tech magazines are that feature them. Tioaeu8943 (talk) 19:00, 11 September 2026 (UTC)reply
In what regard? The bare minimum should not be enough to justify its existence without any substantial coverage or evidence, which this article is severely lacking. Most of the article reads like a collection of trivial material, editorialized statements, and over-glorified facts to promote the company and its products, none of which are notable on their own and would be WP:INHERITED notability, which is not how notability is determined for related topics. Nothing demonstrates that the company itself yields any notability independent of its products being made from some notable systems. It all around reads like a press release. —Trailblazer101🔥 (discuss · contribs)19:51, 14 September 2026 (UTC)reply
Keep. The company appears to have received coverage in third-party publications, including coverage of its products. The current article may need substantial cleanup, better sourcing, and removal of promotional material, but those are editorial issues. The absence of readily available English-language sources does not by itself establish that the company lacks notability, particularly for a Portuguese company where relevant coverage may be in Portuguese.--Leuvraethi mer Dorus (talk) 08:28, 21 September 2026 (UTC)reply
It shouldn't matter that there's a paywall around a source, it should be accessible through Wikipedia Library if someone wants to check. For other sources providing SIGCOV, I'll point to their activities and positions getting regular coverage in mainstream news publications; sample below:
probably LLM output. checking all the sources all but one did exist but none appeared to be actually relevant to the sentence the sources were citing daysantdevtalk19:32, 24 August 2026 (UTC)reply
Keep Did you perform a brief google search first before nominating? The first result is this . While it's not significant, it's something to improve the existing sources. HOWEVER, there's not much more reliable sources to find unless we find it in their language, which might be a pain. For now, it's notable enough to keep the article. ~ Elisapoly (talk) 19:26, 29 August 2026 (UTC)reply
Redirect to Subotica#Infrastructure per WP:NEWLLM. Wikipedia.gptzero.me says with 100% probability that the page was generated by AI. In principle, this could be stubified, but the proposed target already contains basic information about the station. Practice regarding LLM created articles seems to be that we prefer removal over editing. Kelob2678 (talk) 20:00, 7 September 2026 (UTC)reply
Redirect to the new high-speed rail line to Hungary as described in the article, that has lots of coverage. This appears to be an historic building, but both the Serbian and Hungarian wiki articles either have no sourcing, or just timetable listings confirming a station exists... There probably are old articles about the structure itself, but none that I can find. Oaktree b (talk) 18:08, 8 September 2026 (UTC)reply
Relisted to generate a more thorough discussion and clearer consensus. Relisting comment: The consensus is to redirect, but where? Please add new comments below this notice. Thanks, Dclemens1971 (talk) 01:24, 16 September 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
