Wikipedia:Education noticeboard

Welcome to the education noticeboard

Purpose of this page			Using this page
This page is for discussion related to student assignments and the Wikipedia Education Program. Please feel free to post, whether you're from a class, a potential class, or if you're a Wikipedia editor. Topics for this board might include: Content issues created by real or potential student assignments Improving student assignment output quality per the RfC Questions about the trainings or any question from instructors or students Discussion of the Wiki Education Foundation or the Wikipedia Education Program. Students who have done great work and need {{The Excellent New Editor's Barnstar}} Of course, we should remain civil towards all participants and assume good faith. There are other pages more appropriate for dealing with certain specific issues: An issue requiring immediate administrator attention → Administrators' noticeboard/Incidents Vandalism by non-participating students from schools → administrator intervention against vandalism. Cut-and-paste moves/history merge → the cut-and-paste move repair holding pen. Students revealing sensitive personal information → requests for oversight. Please DO NOT make such requests here; reports here are visible to everyone. Course announcements from dashboard.wikiedu.org / outreachdashboard.wmflabs.org → /Wiki Ed course submissions Issues related to dashboard.wikedu.org / outreachdashboard.wmflabs.org → meta:Talk:Programs & Events Dashboard For a list of topics that students should avoid editing → Wikipedia:General sanctions#Active sanctions			Start a new discussion thread. Use an informative title: ==Informative title==. If a thread is related to an ongoing discussion, consider placing it under a level-3 heading within that existing discussion. You should generally notify any user who is the subject of a discussion. You may use {{ping}} to do so, or simply link their username when you post your comment. It is not required to contact students when their edits are only being discussed in the context of a class-wide problem. If no comments have been made within 30 days, your post and any responses will be automatically archived. Please sign all contributions, using four tilde characters "~~~~". If discussion is already ongoing elsewhere or if there is a more natural location for a discussion, please continue the discussion there, and put a short note with a link to the relevant location on this page. If you cannot edit this page because it is protected, please place your comments on this page and they will be addressed. Managing threads If you'd like to make sure a thread does not get archived automatically after 30 days, use {{Do not archive until}} at the top of the section. Use {{User:ClueBot III/ArchiveNow}} within a section to have it archived (more or less) immediately. A brief Archives page lists them with the years in which those now inactive discussions took place.

Click here to start a new discussion

Should WikiEd be terminated?

Bluntly there are too many problems with bad student edits not being reviewed by the class teachers or the associated WikiEd. While a small number are good (I once helped a student with an article that ended up as GA), far too many are bad. As examples you can look at the recent discussion at Wikipedia:Articles for deletion/Community composting, Draft talk:Edwin Samuel Lennox or some of the recent edits by students of Wikipedia:Wiki_Ed/York_University/BIOL_4095_3.0_Applied_Plant_Ecology_(Winter_2026); there are too many other instances.

1. Why are novices in charge of teaching classes?
2. What requirements and training is their for advisors?
3. What reviewing is done by the class mentors, as against expecting others to to do it?

As someone who taught at an R1 university for many years, I always accepted responsibility for my class students and PhDs. Currently WikiEd is a drain on reviewers.

Maybe this needs to go to an RfC. Ldm1954 (talk) 09:12, 1 April 2026 (UTC)

I have had some years of experience with student editors. Problems left by bad edits are rarely fixed by the course teachers and are left for other, like me, to fix. Graham Beards (talk) 10:09, 1 April 2026 (UTC)

I have only 2+ years of experience with bad student editors. When I care about the topic I end up checking the class page then adding all the articles to my watch list and then monitoring them. The most common actions are draftify or revert. The "WikiEd" advisers rarely do anything including not responding to pings. Ldm1954 (talk) 22:41, 1 April 2026 (UTC)

@Ldm1954, in my experience the WikiEd advisers are quite good about responding and getting in touch with professors whose classes are causing problems. But they have to know there's a problem in order to solve it. I notice that this is your first post on this noticeboard. If you come here when you see things starting to go wrong with a class, other editors, including the WikiEd employees responsible for the class, can intervene. -- asilvering (talk) 23:15, 1 April 2026 (UTC)

This may be my first post here, it is not my first ping of WikiEd employees about their classes. My experience over the last years is the same as @Graham Beards and many others (e.g. the AfD discussion and the comments below by @Reywas92). Contrary to the statements by @LEvalyn I have AfD'd, PROD'd or draftified multiple pages or reverted multiple edits for STEM classes. Ldm1954 (talk) 23:27, 1 April 2026 (UTC)

I, too, have AfD'd several WikiEdu-related articles. But "several" is very far indeed from "7,500-10,000". -- asilvering (talk) 23:34, 1 April 2026 (UTC)

I cannot sort my past AfD's on classes, but I think my draftification rate when new articles are dumped in main by classes is at least 25%. Typically they abandon the draft. Reversion of their edits as violating NPOV, unsourced etc is also about 25%. Of course I only do this in STEM, maybe non-STEM is better. Ldm1954 (talk) 23:38, 1 April 2026 (UTC)

Copying my comment from Wikipedia:Village pump (policy)#Student edits: As a start, Wiki Ed classes should prohibit students from creating new articles. New student editors frequently create articles that are not appropriate for standalone pages as they lack notability or are hyperspecific topics. They are very often written as essays with low-quality writing rather than in an encyclopedic style. They often include plagiarism, off-topic content, and other issues you would expect from inexperienced students being forced to do an assignment rather than volunteering like experienced editors who start with smaller edits at a time. I have see far too many AFDs of student articles that should have come nowhere near mainspace had a responsible instructor or an established editor reviewed them first. Some AFDs of problematic articles include Wikipedia:Articles for deletion/Community composting, Wikipedia:Articles for deletion/Systemic injustice in literature, Wikipedia:Articles for deletion/Impact of sand loss on sea turtles, Wikipedia:Articles for deletion/Food waste in Barcelona (see Wikipedia:Wiki Ed/IES Abroad Barcelona/The Climate Crisis - Global Perspective, Mediterranean Context (Spring) and the instructor's other classes for how poor this instructor was at letting his students pick topics), Wikipedia:Articles for deletion/Environmental sustainability of vintage fashion, and Wikipedia:Articles for deletion/Native Americans and horses. Edits to existing articles are also often quite poor as well and it would be great if WikiEd required teachers to review students' draft/userspace edits before they go live. Students usually each have their own topics, reviewed by just one of their peers, when they should really be working as teams on topics.
I would certainly support an RFC to instititute requirements that student work get additional review before going into mainspace. Too many instructors haven't contributed to Wikipedia themselves and have no idea how to ensure work meet our notability, style, and quality standards. I'm astonished how bad many students are and I always wonder what kind of grades they get, as some certainly don't deserve a good one. — Reywas92^Talk 14:22, 1 April 2026 (UTC)

I see criticisms of the entire WikiEd program appear twice a year as the spring and fall semesters conclude, generally prompted by one or two classes that have failed to live up to WikiEd ideals. Indeed, the last such was mid November, at roughly this point in the fall semester. I think that conversation is worth a read. I'd highlight my argument that there is a major confirmation bias problem here: editors will never notice the harmless students, and may not realise the positive ones are students at all, so it will seem that all student editors are disruptive. But WikiEdu interacts with 300-400 classes a semester, and we do not get 7,500 to 10,000 AfDs for problematic student articles. ~ le 🌸 valyn (talk) 23:21, 1 April 2026 (UTC)

I certainly don't think all student editors are disruptive, but I will regularly look at class pages and students' contribution history and also see that many students never actually get their assignments out of draft, or they end up only contributing a few sentences and sources. Many others are simply reverted, without requiring a formal discussion to delete, as most students do in fact choose existing articles rather than new ones. But the bad is often very bad. I recently reverted this awful edit. Luckily I could do that without an AFD but jeez. I wouldn't exactly agree with the OP's suggestion for termination, but we absolutely need better guidelines that instructors actually be experienced with the editing process, encyclopedic style, and key guidelines including notability themselves, and that students' work is better reviewed before it goes live. — Reywas92^Talk 04:51, 2 April 2026 (UTC)

Two thoughts: (1) Terminating WikiEd won't terminate irresponsible instructors who bring their students here. It will just remove something we have that can help get them under control. (Nothing's perfect, but WikiEd is actually pretty good.) (2) Never hesitate to revert. --Tryptofish (talk) 23:48, 1 April 2026 (UTC)

I've seen plenty of good student edit projects, but they really require a teacher who has an understanding of how Wikipedia functions. Inside my frequent topic area of plant science, I've seen various improvements over the years. Such as with the Miss Kim lilac, which went from this to its current state thanks to the efforts of student editor User:Plantlady890. Silverseren^C 23:52, 1 April 2026 (UTC)

The case for WikiEd's existence is that instructors will likely assign Wikipedia editing regardless, so some organizational structure is better than none. At least we have someone to contact if a class goes off the rails, instead of editors trying to cold-contact adjunct professors to ask if they're running a course. In my experience, I've gotten consistently good help from Brianda whenever I've pinged them about a problem.

That said, I do enough NPP to deeply sympathize with the frustration here. Finding the majority of a given class contributing mostly AI slop, copyvio, and non-neutral OR is very disheartening. I think the quality variance traces back largely to instructors. A non-constructive student editor is at least a contained problem: they almost always disappear after the semester ends. A poor instructor compounds and spreads damage over time. Some simply do not have the competence to be driving a bus full of other novices into mainspace, and that seems like a pattern WikiEd may be positioned to identify and address. Zzz plant (talk) 23:55, 1 April 2026 (UTC)

It still surprises me that we don't seem to have any process for stepping in when educational projects go off the rails, especially those which haven't been formally declared by their tutor.

On the 15-minute city article we've currently got three apparently student editors adding the same kinds of inappropriate and sometimes AI-generated material, a lot of which has been reverted. Beyond a couple of welcome templates, it seems this is expected to be handled by existing systems that would stop any three humans from making bad edits: discussion, escalating warnings, and (if the students haven't moved on by then) eventual blocks or article protection.

When we know that it's an educational situation where those three humans are all following the same poorly-considered instructions from a tutor, we might benefit from a process for shortcutting that and going straight to "your school project is breaking Wikipedia guidelines, please stop working on it and inform your tutor of this message" notices and page protection. Belbury (talk) 08:45, 2 April 2026 (UTC)

@Belbury, I don't understand what you're looking to change here? If you report those projects to this board or to ANI, admins will happily protect the page so that students have to go through edit requests. I've done that for this one just now. You just need to ask. Admins, like WikiEdu, can't do anything about problems if no one bothers to report them. -- asilvering (talk) 16:30, 2 April 2026 (UTC)

Also, {{AINBA}} allows you to ping admins wanting to deal with AI cleanup! Chaotic Enby (talk · contribs) 16:36, 2 April 2026 (UTC)

Thanks for protecting the article. What I'd look to change would be guidance and warning templates for average editors in handling that. I hadn't requested article protection because I wasn't sure that it was warranted yet, and whether (unlike a regular protection request) it would be fair to factor in there maybe being twenty other students poised to make the same kinds of edits to the same article. I haven't given the students a voice-of-Wikipedia instruction/permission to stop working on their apparently guideline-breaking and undeclared school project because I don't feel that that's my call to make. Belbury (talk) 17:19, 2 April 2026 (UTC)

@Belbury, no need to worry about "voice-of-Wikipedia". Really the best thing in these circumstances is for students to hear from Wikipedians. If you don't want to get into an extended back-and-forth with them, though, you can always drop something short like "Hi, it looks like you're editing as part of a student project. Can you please tell your instructor to make a post at WP:EDUN introducing themselves? Thank you!" -- asilvering (talk) 17:45, 2 April 2026 (UTC)

I'd more or less done that, asking them and their instructor to read WP:ASSIGN. (This had no effect.) It wouldn't have occurred to me to suggest their instructor also posts at WP:EDUN.

If badly-managed student projects are a problem, I think it would help to have some clear guidelines on how to handle them efficiently, a canonical talk page template for editors who don't want to open a lot of back-and-forths with a full class of students, and to decide at what point (if any) Wikipedia considers it appropriate to close a class project down and explicitly tell the students to stop working on it. Belbury (talk) 18:04, 2 April 2026 (UTC)

If you warn and it has no effect, you're always welcome to head to ANI for immediate admin action. For complicated stuff, this noticeboard is a good place. -- asilvering (talk) 20:38, 2 April 2026 (UTC)

It looks like this is a perennial issue. I'd love to see some data on if WikiEd editors overall tend to produce poor quality edits, or if teachers respond to issues in a timely manner. Closing the whole program seems extreme unless there is documented evidence of extensive work needed to fix bad edits. SenshiSun (talk) 02:02, 2 April 2026 (UTC)

I agree about data being useful. As one current example of data, I just found two newly created articles in main from Wikipedia:Wiki_Ed/Jackson_State_University/U.S._History_II_(Spring_2026). One is a US patent, the other is the person who filed it. Neither has any SIGCOV, both were already tagged as AI, and are now draftified (Draft:Virgie Ammons and Draft:U.S. Patent No. 3,908,633). Since they are the only new pages for that class it is currently a 100% failure rate. (For reference, there are 2 other pages where substantive edits have been made, and on one most have been previously reverted by a different editor.) The responsible WikiEd is Brianda (Wiki Ed), the course instructor is User:Jbhistory who has 9 edits. Ldm1954 (talk) 02:49, 2 April 2026 (UTC)

I agree with Silverseren. It really depends on whether the instructor is knowledgeable enough (on wiki) to be on top of the student assignments. I'm running a course right now (see above) and nobody has complained about my students even though climate change topic was just delisted a month ago from contentious topic designation. Unlike creating a new article, I go with something small (just one or two edits) so that it's easier to manage. I already did a trial run last summer by asking students to write as if they would be submitting the change, but send it to me by email instead of actually saving the edit. Once the method proves to work, I carry it out this term with actual on-wiki edits. This starts off from screening the topic that students will be writing about (which includes rejecting if it's inappropriate or if I think the page is already in a good condition that students are unlikely to bring a lot of value with their edit). And when they are done with directly editing in mainspace, I don't delegate the review process to another classmate which sometimes has the tendency to rubberstamp their peer's work and not fully critique what's wrong with the writing. I go in, check for close paraphrasing and fix stuff like reference formatting and moving contents to more appropriate page. I stayed on top of the editing so that the issues don't snowball and overwhelm the instructor. I used WikiEd's training module which is beneficial for students to learn about how to add reference. As a counterpoint, I have seen instructors who didn't properly vet student's contributions and let glaring issues slip by. I don't think we can fault WikiEd for this because WikiEd can't control what the instructor does or doesn't do. OhanaUnited^{Talk page} 06:03, 2 April 2026 (UTC)

Comment I think it is worth looking at the instructor cleanup instructions. It includes the specific, and to me troubling statement:

While it's not required for you to fix your students' contributions yourself, we'd love for you to give it a try.

If you follow the links you will see that the current training page for instructors does not require any Wikipedia expertise.

I think we have to change this. While it might have been relevant in the early days, I feel we have to require much more from the instructors if they want to use our time to help their courses. Remember that teaching is a requirement at universities, and nominally what the pay is for. The time of Wikipedia volunteers is subsidising these classes. At the very least instructors should be Extended Confirmed Users, although this seems too low to me.

Similarly the WikiEd experts should have at least the expertise of a WP:NPP, and probably the same rights including tagging and draftification. I think if they spent some time in the dregs and chaos of NPP they might have a much better feel and be better equipped to handle problems. (I have not found details of what training is required to be a WikiEd expert.) Ldm1954 (talk) 16:14, 2 April 2026 (UTC)

That line from the training for instructors makes me cringe. That's something that would clearly benefit from revision by WikiEd. The alternative would be that the community might enact a policy that requires instructors to do this or get blocked, which would certainly lead to a lot of drama that would be best avoided. --Tryptofish (talk) 22:21, 2 April 2026 (UTC)

Or the teacher would cheerfully get themselves blocked, so they can say "See? Now I can't, so I don't have to." WhatamIdoing (talk) 00:45, 3 April 2026 (UTC)

I feel like if you're a teacher and you get yourself blocked you probably shouldn't be running WikiEd... Sesquilinear (talk) 02:02, 13 April 2026 (UTC)

There are requirements that WikiEd has in place for the classes it supports, and then there are suggestions. The reason we don't see many classes with like 300 literature students editing medical articles anymore is because WikiEd said "hey we have this great infrastructure you can use, but you have to agree not to do these things that make a mess of Wikipedia". It's also that, and not the phenomenon of instructors having students edit Wikipedia, which you're proposing to "terminate" here. Big issues with classes have gone way, way, down since the pre-WikiEd era (and its early days, before those guidelines/best practices/training modules were fully developed). Long-timers may remember the copyright issues from the Indian program, the psychology and medicine blow-ups, etc. from 2012-2016. These days, student editors make edits that are a bit better than a typical newbie, and come with the bonus of having multiple people you can ping for help. Those people might not always respond immediately, but it's better than the zero people you can ping for help with a random new user. If you make the requirements for instructors to teach above board with WikiEd too onerous, and I'd argue requiring extended confirmed is a totally unrealistic bar to set, the risk is that, at least for some nontrivial subset, you're just turning known students with a support infrastructure into random users with no support infrastructure. If someone is a Wikipedia expert, they don't really need the infrastructure that much, after all. But I agree data is lacking. WikiEd (and not even including all the students below the radar) account for about a fifth of all new active editors on enwiki -- thousands of students editing thousands of articles in hundreds of classes every semester. That is in part why "look at this one class" or "I've found 30 bad articles over the last few years" just aren't that compelling as evidence. It's a good reason to get WikiEd staff involved and have words with the professor (and indeed nobody should be affording students any extra leeway than a typical editor), but it's poor evidence of a systemic problem. On the other hand, I do miss the days when WikiEd posted regular reports bragging about the many good articles students did create. Also not a systematic review, of course, but at least provides a counterpoint to the "I found a few bad articles and/or a troublesome class, so let's kill it for everyone" threads that pop up here once a semester. (disclosure: I'm not uninvolved - I was involved with WikiEd some years back, and included Wikipedia in my classes in the pre-WikiEd era [mostly without incident :) ]). — Rhododendrites ^talk \\ 16:41, 2 April 2026 (UTC)

I am very confused by your statement that ECU is too high, that criterion is "the account has existed for at least 30 days and has made at least 500 edits". This is really minor.

I also have seen far more than "30 bad articles over the last few years". I would say that of the NPP reviewing I have done in STEM, the number of student articles which require major repair is about 25%. Yes, there are exceptions, and in the past I had graduate students create pages as term pages, all of these are still in existence; I checked them personally. However the problem is real. Typically experienced reviewers in Chemistry/Physics (where I mainly work) repair the problems via PROD/AfD/AfC/Merge etc, they do not raise them here. Classes come up continually on the relevant project talk pages, for instance Wikipedia talk:WikiProject Chemistry/Archive 40#Essays for a class project,

Wikipedia talk:WikiProject Chemistry/Archive 37#Seeking input on a student handbook for chemistry and

Wikipedia talk:WikiProject Chemistry/Archive 38#Student projects from UCLA Ldm1954 (talk) 17:16, 2 April 2026 (UTC)

@Ldm1954, extended-confirmed is not "minor". Less than 1% of contributors ever get ECR, and of those few who do, it usually takes them more than a year to reach that point. WhatamIdoing (talk) 01:00, 3 April 2026 (UTC)

I am OK if we reduce the requirement, but there should be something. At least when I was active we never asked an economics professor to teach quantum physics, we always required competence, preferably at the research level, particularly for more advanced classes. I am definitely troubled by the prof in charge having, for instance, 9 edits. Ldm1954 (talk) 14:33, 3 April 2026 (UTC)

I don't know if a blanket rule makes sense. We need someone to know what they're doing, but it doesn't necessarily have to be the instructor. The Wikipedia portion of a class might, in practice but not necessarily visible to us, be led by a Teaching assistant who actually knows what they're doing. WhatamIdoing (talk) 17:31, 3 April 2026 (UTC)

In every course page I know of TA emails are included, along with office hours and sometimes other social media handles; I have never seen any of this on WikiEd. I have also never seen any TA comments on any talk page. The customers (the paying students) are peer reviewing each other.

From what I have seen the students are poorly trained, and I doubt that there is any record that they actually did the WikiEd training. I am required to complete various safety training modules with quite tough questions; an ex-postdoc got fired from a DOE lab for not completing safety and ethics training. Similarly our undergrads are required to take certain safety training, the grads and postdocs far more. I suggest trying one or two of those at https://dashboard.wikiedu.org/training yourself to see how they compare to required academic/company/lab training. Ldm1954 (talk) 17:52, 3 April 2026 (UTC)

Maybe you should volunteer to help with one of these classes, so you'd know what you're talking about. WhatamIdoing (talk) 20:18, 3 April 2026 (UTC)

We used to have regional ambassadors, including myself, who are experienced Wikipedians and act as the go-between person between the instructor and the community, online ambassadors who answered specific on-wiki questions and campus ambassadors who offered hands-on training on campus. While the outreach page still displays the ambassador program, I'm unsure if the program is still active. As for extended confirmed user, that is an insanely high bar to demand. And I don't understand why we're bringing up discussion around safety training modules. Seems like a lot of these points are simply distraction. OhanaUnited^{Talk page} 19:18, 6 April 2026 (UTC)

As a clarification about safety modules, that was comparing the existing WikiEd training modules to standard training module that students in the sciences are required to take.

As previously stated, I have no problem with a lower level of expertise expected of the instructor. However, we don't ask a genetics professor to teach a class on Dante’s Inferno. Ldm1954 (talk) 19:28, 6 April 2026 (UTC)

Do you have any recent evidence that a professor is assigning students to write topics that are outside of their subject expertise or is this vague hand-waving conjecture? For the BIOL 4095 3.0 Applied Plant Ecology course at York University example that you initially raised, I inspected the entire assignment list since I took conservation ecology for my bachelors and currently working in a position related to one health so I'm in a qualified position to judge. Out of 97 articles listed, to your credit, there are perhaps 3-4 articles that are off-topic (some are borderline like Edward Burtynsky, and some are obviously off-topic like Opium Wars) that I'm not certain if the instructor can meaningfully evaluate off-topic contributions. But this course isn't doing anything much different than what my course is doing right now (asking students to add content and add references). I had to self-teach myself on sudden ionospheric disturbance to evaluate my student's edit. If the best examples you can come up with are from 2015 or 2016 then I don't know why you're so fixated to nitpick every minor blunder that students or instructor make. OhanaUnited^{Talk page} 21:32, 6 April 2026 (UTC)

I did not say anything about the the topics, I have never questioned the competence of the professors in their areas. The point I have raised several times is the Wikipedia expertise. Having an instructor teach a class where an integral part is writing or editing a Wikipedia page, when there is zero requirement that the instructor has themselves ever done that is. IMHO, problematic.

Please do not keep using the the word you which implies that I am the only editor who has ever had to repair student editing problems. And also please note that I stated right at the front that sometimes the pages are good. An example, since specifics seem to be needed, is Computational chemistry which is both GA and high-importance. After the class the student put in extensive time to work on the page, and it has survived for 2 years as GA (with some additions from various people, which is good). Ldm1954 (talk) 22:01, 6 April 2026 (UTC)

Excuse me? I have cleaned up my student's editing problems, including minor ones like these to major ones like these . I have also participated in the cleanup of other classes, including deleting a userpage with reference misrepresentation, as early as 2014. So I politely ask you to stop making stuff up. It's clear, from multiple editors' comments, that you don't know what you're talking about. OhanaUnited^{Talk page} 20:34, 11 April 2026 (UTC)

I doubt that there is any record that they actually did the WikiEd training - Again, if you just did some looking around and the systems that you're advocating to shut down, you would see there is. Open any class page and mouse over the "student editors" at the top to see how many students are up to date on trainings. IIRC professors have a more detailed breakdown of who's done what that isn't public. — Rhododendrites ^talk \\ 20:45, 6 April 2026 (UTC)

• Support allowing WikiEd to exist Previous coverage - Changing the face of Wikipedia, August 2021.

1. Two possible alternatives are 1) replacing the program with nothing and 2) replacing the program with anything else tried or proposed. I feel that of the Wikimedia Foundation programs, Wiki Education is the most successful, at the lowest cost, by any measure anyone has used for analysis and comparison.
2. WikiEd's mistakes may be visible and detectable, but by all measures which anyone has ever devised, Wiki Ed overall has a lower problem rate than 1) any other Wikimedia outreach campaign 2) general editors who come to Wikipedia without a campaign affiliation. Regardless of whether Wiki Ed is good, it is the best we have ever developed in terms of getting edits with a low rate of problems.
3. WikiEd has been more successful and less expensive in all our success metrics as compared to other WMF funded and encouraged programs. There are reasons why we want a diversity of programs, but if the goal were to reduce errors and increase our success metrics, then we could achieve that by cutting every other campaign we have and only funding Wiki Ed.

All of our programs have problems, but there is consensus that doing programs gives much better outcomes than not having any outreach programs. The Wikimedia Foundation has a strategic role in recognizing when to drop funding to address which problems, in which order. If there is anyone here who wants to pull levers on what the success metrics are and what gets funded, then contact the WMF/ volunteer collaborative team for North America at meta:Grants:Regions/North_America.

Through the WMF's process, it is currently decided that Wiki Ed's problems are a good balance for cost / benefit as compared to every other program where we spend money. Wiki Ed's grant from WMF is small considering that 20% of English Wikipedia accounts get registered through this process, and they have the highest rate of good citation-backed edits with the fewest reversions.

Please reconsider using negativity about Wiki Ed. Their current WMF grant is about $200,000/year in the context of WMF having an annual spend of about $200,000,000. And I highly encourage anyone who cares to participate in the grant discussions - it is unusual and powerful for any organization to take user input in its grants funding process and few people participate in the money discussions. If anyone here would actually join the money conversations then there are so many people who would invite you, meet you, talk to you about whatever you wanted to know. Go up a level from Wiki Ed and look at the other millions of dollars to understand a bigger picture. Bluerasberry (talk) 16:48, 2 April 2026 (UTC)

A Signpost story from August 2021, before the era of chatbot slop, necessarily fails to address the escalating problem we have on our hands now. Universities are literally paying for enterprise access to the slop machine. What are we going to do about students who are primed to edit in a way that is antithetical to our goals? I am sympathetic to the argument that we need something like Wiki Ed because any support infrastructure is better than none at all, but it also looks like we need to draw brighter lines for instructors.

One recent explanation of Wikipedia for outside experts had this to say: One strategy that we wish to caution against is assigning students the task of editing Wikipedia, at least until you’ve been a Wikipedia editor yourself for long enough to know what can go wrong. Over the years, many schools have tried involving their students with Wikipedia. This has had mixed results, and many long-time Wikipedia editors have negative reactions to the idea thanks to its being most noticeable when it goes badly. A student with a shaky grasp of the material, writing out of obligation, without the experience to tell a good source from a bad one, and bringing all the habits acquired from essays and term papers ... we shudder a little just to think of it. Stepwise Continuous Dysfunction (talk) 21:07, 2 April 2026 (UTC)

WikiEdu has done a very good job on the AI front. Far better than anyone could have hoped, honestly. I wish I could find the article that broke down some numbers on the topic, but I can at least refer you to . -- asilvering (talk) 21:13, 2 April 2026 (UTC)

It's here. Students appear to be the only group of new users whose contributions are systematically evaluated for LLM-generated content now. — Rhododendrites ^talk \\ 21:16, 2 April 2026 (UTC)

Is that true? I've seen a lot of rejections recently. Deb (talk) 13:16, 5 April 2026 (UTC)

I don't know what the current state is, but I know that Wiki Edu was mass-screening contributions from students for LLM use. See https://wikiedu.org/blog/2026/01/29/generative-ai-and-wikipedia-editing-what-we-learned-in-2025/ for more information. WhatamIdoing (talk) 16:36, 5 April 2026 (UTC)

There are still issues and staff are still reverting a large number of articles for tripping that conservative screening model that they use. Superb Owl (talk) 21:50, 20 May 2026 (UTC)

Seconding @Bluerasberry's point about joining the regional grants committees. North America especially needs more volunteers for this work. But yes this is where evaluation of Wiki Ed and their impact on en-wiki happens and where they apply for funding. They are one of the most successful grantees for a reason - they have the most impact of any organization , at least that's funded through North America region. I spent 4 years on that committee and saw their consistent results/reporting year after year. Matthewvetter (talk) 13:47, 8 April 2026 (UTC)

I disagree - I have seen very little evidence of true impact so far and would love to see it if you know where that information is. The impact I see is around superficial growth metrics like # of edits and how important the topics are that they are editing. There seems to be a focus on doing enough to win foundation grants and showing growth but not enough of a focus on quality. I have started a page called Help:Wiki Education to document discussions like these, resources, questions from Wikipedia editors and suggestions for improvement. Superb Owl (talk) 21:49, 20 May 2026 (UTC)

@Superb Owl, I'm not sure why you've placed that in the help namespace? -- asilvering (talk) 22:10, 20 May 2026 (UTC)

I'm boycotting the Wikipedia namespace until we get Visual Editor (unless there is a consensus to move a particular article there) Superb Owl (talk) 22:18, 20 May 2026 (UTC)

I have a serious concern that this new page is competing against WP:ASSIGN. --Tryptofish (talk) 22:21, 20 May 2026 (UTC)

Did not know about that! Thanks for passing along and I'll start linking to that page and removing redundancies Superb Owl (talk) 22:42, 20 May 2026 (UTC)

Superb Owl, boycott not necessary. Try this link, for example. Mathglot (talk) 19:20, 27 May 2026 (UTC)

• I'll echo what Rhododendrites and others have stated. This is a necessary program. Getting rid of Wiki Education will not solve the problem. For full disclosure, I am a former Wiki Education employee as well. This would have fairly disastrous results. For one, people would still use Wikipedia in their assignments. They aren't going to stop because Wiki Education no longer exists. You would just have a class that you may not be able to identify. Secondly, this would eliminate the course's ability to have a dedicated person assist them. They would have to rely on volunteer editors, who would in turn be limited by their off-Wikipedia lives. They could be helping and then just disappear off Wikipedia for months (or forever) because of said IRL obligations. Having a staff whose literal job centers upon helping students and educators helps relieve things for volunteers because we have someone who will reach out to the students/educators. If one of them gets sick, then another employee steps in. Third, WE does actually try to step in and curtail any problematic editing. When I worked with them, a large part of my job was reviewing student work and trying to step in when needed. Of course it's more difficult when you have a huge amount of courses, but trust me when I say that we were as vigilant as possible. From what I've seen, the current staff is doing the same thing. Finally, I'm a little afraid of how shutting down Wiki Education would be perceived, as it would likely come across as the site telling the educational community that they are unwelcome. Those who are anti-Wikipedia would likely have a field day with this. ReaderofthePack(formerly Tokyogirl79) (｡◕‿◕｡) 19:23, 27 April 2026 (UTC)
• (new here) While I understand the concerns, and share some of them, I haven't seen any convincing argument that (1) we would be better off without WikiEd; (2) WikiEd student contributions are a net negative; or that (3) WikiEd student contributions are worse than those from other loose or formal groups of new editors such as rabid fandoms and people attracted to high profile or controversial topics. My first real engagement with WikiEd revealed a useful framework for dealing with potentially problematic student editors with support from responsive staff members. If only every new editor came with this much support! It would be great to see increased coordination with WikiProjects or other collaborations on key content areas but I recognize this would require a lot of work from volunteer editors so may not be feasible. Ultimately, we can't stop faculty from assigning these projects and having a support system and easy ways to identify and track these contributions is to everyone's benefit. In fairness to those who support eliminating or reigning in WikiEd, I have not had the experience of repeatedly editing in content areas that are 'targets' for these assignments. —Myceteae‍🍄‍🟫 (talk) 00:55, 21 May 2026 (UTC)

  Really what we are arguing about is whether Wiki Education should be pushing for growth (quantity) or slow down and focus on improving the quality of the contributions and hold instructors to higher standards. I support Wiki Ed slowing down and retooling all of their metrics to focus on quality instead of quantity.

  I also might be more frustrated than most since Wiki Education has made a push into politics, which overlaps with some of the areas I edit, and have reverted 85% of the content added over 3 articles in the past month. What's left is ok but not worth the hours it took to distill those edits down to something usable. Superb Owl (talk) 06:40, 21 May 2026 (UTC)

  So far this calendar year, Wiki Edu has supervised more than 300 classes, in which 6,300 new editors have contributed to almost 6,700 articles, including adding almost 50,000 (fifty thousand!) sources.

  So far this year, you've had problems with most (but not all) of the edits to three (3) of those 6,700 articles.

  You propose them "slowing down". They already spend weeks and months training, whereas the typical newbie only edits for one or two days. What does "slow down" look like to you?

  You propose them "retooling all of their metrics to focus on quality". They are already training new editors (better than we are). They are already checking for LLM content (better than we are). They are already tracking the addition of citations (better than we are). They already tell students to double-check each others' work before adding it to an article (far more pre-edit review than our usual). We already know that their students' revert rates are lower than non-student editors. Students are massively (like three orders of magnitude) less likely to get blocked. So all of this "focus on quality" is already happening, but you want something more. What more do you want? WhatamIdoing (talk) 20:45, 21 May 2026 (UTC)

  @WhatamIdoing, in short, moving personnel away from recruitment and towards quality control would be welcome. Whether that's better training materials or actually reading what their students are writing before they add it to Wikipedia. Not all instructors do this, but I think they should. Perhaps Wiki Education needs to hire seasoned Wikipedia editors at the end of each semester to help respond quickly to the onslaught of submissions (especially those from first-time instructors or instructors whose classes have not had a great track record). For more specific proposals, please see: Help:Wiki Education. And if you have access to the data on those statistics, I would be curious to see it. Superb Owl (talk) 20:59, 21 May 2026 (UTC)

  @Superb Owl, my understanding (WikiEdu folks, please correct me if I am wrong) is that WikiEdu isn't exactly going around trying to recruit new classes. They're contacted for help by the instructors, or sometimes they're called in when a non-Edu class project on Wikipedia is going poorly and we'd like someone with experience to help out. In solidarity, asilvering (talk) 21:21, 21 May 2026 (UTC)

  @Asilvering, that's a good question but was not the impression I got from reading their Annual report for 2025/26 and seeing their dashboard where all of their metrics also suggest a celebration of growth, which is great so long as there aren't metrics that capture the quality of contributions that might lead to better outcomes in terms of the student and instructor experience as well as the long-term editor experience. Superb Owl (talk) 21:37, 21 May 2026 (UTC)

  But WikiEd is the people hired to respond quickly to new submissions and undertake cleanup. ~ le 🌸 valyn (talk) 21:22, 21 May 2026 (UTC) -- Like, that's the noticeboard you're at right now. The noticeboard to take a student-created problem, and make it their problem. ~ le 🌸 valyn (talk) 21:25, 21 May 2026 (UTC)

  I think their idea is that Wiki Edu would train people all semester long, and then some additional Wikipedia editors would basically grade the students' work in the last few weeks of the class. WhatamIdoing (talk) 21:24, 21 May 2026 (UTC)

  Do you have any idea how small this organization is? I think they only have one person doing anything that could be called "recruiting". A lot of classes either heard about it by word of mouth, or the instructor has been doing it for years. (The metric I'd like to see is a very low percentage of eligible classes not working with Wiki Edu.)

  This "help" page does not convince me that you have enough knowledge to be providing advice to other people on this subject. For example, you complain about duplicate refs. You give a bold-faced link to a section in a guideline that says "Do not discourage editors, particularly inexperienced ones, from adding duplicate citations when the use of the source is appropriate, because a duplicate is better than no citation". You're basically violating that guideline right there by complaining about it. And then the very next paragraph in that same section in that guideline links to Help:Citation tools#Duplicate reference finders, where you will find multiple answers to your complaint that "Honestly, this is something where we should have a tool to automatically detect and merge these references together". I know that Wikipedia:Nobody reads the directions, but maybe just a little bit more reading of the directions would be appropriate, if you're going to complain about long-term, good-faith efforts to improve Wikipedia? WhatamIdoing (talk) 21:23, 21 May 2026 (UTC)

• Throwing an idea out there... what if instead of extended confirmed or whatever, teachers leading WikiEd classes must write at least one article of their own and get it accepted via AFC. Headbomb {t · c · p · b} 02:10, 7 July 2026 (UTC)

  Are instructors in the WikiEd program the problem? I was under the impression that the problem, to the extent that one exists, is with the courses, instructors, and students who don't work with WikiEd or a similar program in their home country. (And I don't personally think that there is a significant problem - I am asking for the sake of clarity.) ElKevbo (talk) 02:29, 7 July 2026 (UTC)

  @ElKevbo, you're correct. That doesn't mean there aren't problematic WikiEdu classes every so often, but when the ones without WikiEdu support go wrong, it can get quite bad. In solidarity, asilvering (talk) 04:04, 7 July 2026 (UTC)

  It seems both can be problematic. Some of the worst offenders bypass WikiEd (but occasionally, so do some of the best contributions). The problems are often worse and harder to get a handle on when they are done outside the WikiEd structure. As to Headcomb's suggestion, I worry that raising the barrier to entry too much will discourage participation in WikiEd and make the problems worse. Many competent and engaged Wikipedians have never created a new article from scratch and many (most?) assignments don't require students to create one. —Myceteae‍🍄‍🟫 (talk) 15:44, 7 July 2026 (UTC)

  If they don't requite a new article, then I'd suggest the instructor to do the assignment themselves at least once. This way they both have an example to show, and will be able to better guide the students.

  Anyway, that was the idea behind the idea. Headbomb {t · c · p · b} 15:59, 7 July 2026 (UTC)

  I like this as a suggestion. The more engaged and experience the instructor is, the better. —Myceteae‍🍄‍🟫 (talk) 16:56, 7 July 2026 (UTC)

  I don't like it. Perhaps offering instructors a way to, voluntarily, experience editing from their students' perspective could be a useful feature to make available to instructors who actually want to do it. But making a requirement like this for instructors will make most faculty react quite negatively to something they will see as condescending to them, and will drive a lot of class projects to simply take place outside of WikiEd supervision. --Tryptofish (talk) 23:28, 7 July 2026 (UTC)

  To clarify, I also do not support this as a requirement. By I like this as a suggestion I meant offering instructors a way to, voluntarily, experience editing from their students' perspective. I think we should encourage substantive engagement with Wikipedia and giving the specific use case for your assignment a trial run is a good way to do that. And I would acknowledge that even that has serious limitations. An instructor making a handful of contributions and then assigning dozens of students to make superficially similar edits won't have encountered every situation that their students will. Overall, I support keeping the barrier to entry modest so as to make WikiEd inviting. —Myceteae‍🍄‍🟫 (talk) 01:48, 8 July 2026 (UTC)

The premise behind the OP is fundamentally flawed. It is an attribution target error, which starts off correctly identifying Wiki Ed students as the source of many problematic edits without seeing the real cause and then incorrectly identifies Wiki Ed as the cause (or at least, failure to police it). While there are some aspects here of selection bias (as LEvalyn said, "editors will never notice the harmless students"), the real problem is the misattribution of blame. It is not (some subset of) Wiki Edu students per se that are the source of many problematic edits, it is (some subset of) *all* new editors that are the source of many problematic edits. Wiki Edu is close to student editors and clearly correlated with them, but Wiki Ed is not causing (or exacerbating) the problem, it is mitigating the problem. Its visibility and correlation make it a convenient target. Hospitals are not generally the cause of illness (despite nosocomial infection!), they just deal with a lot of ill people.

As an analogy: I see evidence of disruption, personal attacks, copy violations, original research, AI slop, bald COI, unreferenced content, and all sorts of problems all over Wikipedia, but nowhere more than at AN/I and on admin talk pages, or user Talk pages where admins are involved. But that doesn't mean that we should halt WP:RFA and desysop all our admins because evidence of poor behavior is visible like Pig-Pen's stinking trail wherever admins tread.

Wiki Ed is not the cause of problems among student editors, the student editors (like all new editors) are the cause. Wiki Ed mitigates the severity and the scope of the problem without totally wiping it out. New editors, and among them, student editors, will continue to make mistakes and cause problems as long as new editors are allowed to edit Wikipedia. The underlying causal variable is not participation in Wiki Ed but newness to Wikipedia; Wiki Ed merely makes one subset of new editors more visible while attempting to improve their outcomes. Proposing restrictions on Wiki Ed or proposing its dissolution reveals a fundamental misunderstanding of the location of the problem. Mathglot (talk) 17:17, 7 July 2026 (UTC)

I really like Mathglot's analysis here. I also think that there isn't any serious reason to think that WikiEd should be terminated, and that the renewed discussion in this thread that had been going quiet indicates that most thinking about improving things is going in other directions. --Tryptofish (talk) 23:24, 7 July 2026 (UTC)

I just posted about this on Village Pump. Wiki Ed does not work at all in practice. It is a huge detriment to a variety of articles I have seen. It really needs to go. Ilov3gam3z (talk) 17:59, 9 July 2026 (UTC)

@Ilov3gam3z, what does "Wiki Ed" mean to you? Do you mean "university classes should edit Wikipedia with no training, no advice about which subjects might be suitable, no supervision, no encouragement to make test edits in their sandboxes instead of in live articles, and no paid staff person we can complain to when things go wrong"? Or do you mean "new editors are bad, so we should have fewer of them"? Or something else? WhatamIdoing (talk) 18:51, 9 July 2026 (UTC)

??? I mean Wikipedia Education, the people who do the college/university things ofc. New editors are fine. We all were one, after all. Ilov3gam3z (talk) 19:02, 9 July 2026 (UTC)

So you would rather have these classes doing their assignments:

• without anyone training the students on our policies;
• without anyone telling the professors that their planned assignment is bad;
• without anyone telling the professors not to base the assignment grade on whether the edit was reverted;
• without anyone keeping a list of students and checking up on the quality of their edits;
• without anyone telling the students that their edits should be made first in a sandbox; and
• without anyone who is paid to resolve problems created by the students.

Because that is the alternative. There is no fairy godmother to wave a magic wand and make classes disappear. There is only "have a program to reduce and contain the mess" or "have an uncontrolled mess on our hands". We did that last option years ago, and it still happens in non-US/non-Canadian classes. Wiki Ed was created because our experience of the latter option was that it was very bad. WhatamIdoing (talk) 19:16, 9 July 2026 (UTC)

I believe that with the complexity of our policies taken into account, and the fact that more students would only do Wikipedia classes because of Wiki Education, this would not be as dramatic of an effect as you make it out to be. Without knowledge of our policies (which the Wiki Education folks atleast know the very basics I am assuming) most edits done by classes would get taken down for vandalism quickly by RC patrol. Also, less students would do Wikipedia classes without the Wiki Ed and therefor the problem would be reduced. Ilov3gam3z (talk) 19:53, 9 July 2026 (UTC)

Your assumption that if Wiki Ed did not exist, then students would not edit is false.

First we had the students. Then we created Wiki Ed. If we hadn't created Wiki Ed, we'd still have the students. WhatamIdoing (talk) 22:01, 9 July 2026 (UTC)

I'm open to the possibility that Wiki Ed's outreach activities and growth initiatives attract even more classes and students than might show up spontaneously. But even if that is the case, it still comes off as a net positive for the reasons you state and the many other statements here. —Myceteae‍🍄‍🟫 (talk) 22:07, 9 July 2026 (UTC)

• Does anyone have any data on how many WikiEd students continued editing after their assignment? Out of the dozens I’ve seen, not a single one, though maybe some create a different account? There’s just such a big difference between editing as a volunteer and effectively being forced to. Student editors' priority is to pass their assignment, not to improve the encyclopedia. As newbies, they’re usually assigned a high-profile article (already in alright shape), which they are just not capable of improving. I’d be in favour of limiting it to low-traffic or poor-quality articles (where imperfect additions are still improvements), with an emphasis on niches. Kowal2701 (talk, contribs) 21:12, 9 July 2026 (UTC)

  @Kowal2701, there are many editors who have continued to edit after starting for a class assignment, some of whom go on to be really quite dedicated contributors. (You may have heard of Primefac.) You're certainly correct that there's a big difference between doing something as a volunteer and doing it for a grade, but as far as affiliate-sponsored editor-training initiatives go, it's my understanding that WikiEdu is by far and away the most successful. The exact numbers we can't know, for the reason you describe. I would always caution students that they should create a new account if they wanted to do volunteer editing, for their own safety, since WikiEdu classes link your account to some quite specific information (and to other accounts that share those properties). Regarding "usually assigned a high-profile article", I don't think this is at all the case. Most WikiEdu students I come across are not doing so, and WikiEdu does discourage that kind of thing. But just as you can only lead a horse to water, you can't force students (or their instructors) to take good advice. In solidarity, asilvering (talk) 04:12, 10 July 2026 (UTC)

  @Kowal2701, that's a pretty bad narrative. If you head over to Wikipedians stats table, only 30% of all registered users (students AND non-students) made a single edit. This alone already demonstrates WikiEd and participating instructors having a better success rate at converting registered accounts into actual editors no matter how many edits they end up, compared to the "no WikiEd involvement" control group. OhanaUnited^{Talk page} 05:02, 10 July 2026 (UTC)

  Thanks for the replies, I wasn’t aware of that about Primefac, and what WAID said above re WikiEd being a reaction to course modules about Wikipedia makes a lot of sense (I’m sure many like myself enter these discussions with the opposite assumption). Until we can have some data on editor retention, it’s impossible to know WikiEd's value and the tradeoff re disruption/cleanup. The data should exist, I’ve seen some Wikimedia surveys that ask whether you’ve done a WikiEd course or lists it as an option re how you got into editing. Honestly I don’t see the value unless retention rates are decent (compared to non-WikiEd newbies), as terrible as this sounds, newbies are often disruptive initially and we tolerate that because of the chance of retention and their potential, the value to the encyclopedia of a single regular editor is worth a silly amount of account creations, and this needs to be the goal Kowal2701 (talk, contribs) 05:50, 10 July 2026 (UTC)

  I can only offer anecdata, but when I have taught through WikiEd, my students initially have no idea they are allowed to edit Wikipedia, and afterward they universally describe Wikipedia as a welcoming place that is easy to learn about and get involved in. Then they tell their friends! I think that is valuable word-of-mouth to form the top of a potential recruitment 'funnel', and different from what many attempted new editors might say. ~ le 🌸 valyn (talk) 07:04, 10 July 2026 (UTC)

  @Kowal2701, I'm not sure that "retention" is really the appropriate goal to aim for with WikiEdu. Very, very few Wikipedia editors stay for any real length of time, but that doesn't mean they aren't building the encyclopedia in the time they do spend here. While we have some highly prolific editors cranking out GA after GA, most of the encyclopedia has been built in pieces at a time. Lots of people just want to write a single article on a topic we didn't cover; they submit the draft and are never seen from again. If WikiEdu students are more productive than your average newbie, that's still better than baseline even if they leave. In solidarity, asilvering (talk) 07:16, 10 July 2026 (UTC)

  agreed, tbh idek what percentage of newbie edits are constructive (I'd expect 30-50% nowadays, and a big difference once you filter out bad-faith actors?) Kowal2701 (talk, contribs) 07:25, 10 July 2026 (UTC)

  You could ask at WP:RAQ.

  One data point that we sorted earlier is this: Among people who create an account and make an edit in a given year, about one in six get blocked. Among students who create an account and make at least one edit in a given year, about one in 4,000 get blocked. That's a difference of two and a half orders of magnitude. WhatamIdoing (talk) 17:41, 10 July 2026 (UTC)

  I asked about this in 2023, in response to this ENB discussion. This resulted in three Quarry queries being generated, for example, quarry:query/72703: edits by students since 10 months after 'start of course'. (Note: those results took five hours to generate, and the page you are looking at was generated in 2023. Some table changes have occurred since then, and it is possible that rerunning it now might fail.) Mathglot (talk) 19:14, 10 July 2026 (UTC)

  The very first editor on that list now has over 8,000 edits with the most recent edit this month. An outlier to be sure, but quite impressive! —Myceteae‍🍄‍🟫 (talk) 19:41, 10 July 2026 (UTC)

  I wonder whether that could be added to Wikipedia:Student assignments, perhaps in an ==Outcomes== section. WhatamIdoing (talk) 19:58, 10 July 2026 (UTC)

  Done: . --Tryptofish (talk) 00:07, 11 July 2026 (UTC)

  Please monitor WP:RAQ#Wiki Ed student retention (redux). Mathglot (talk) 20:21, 10 July 2026 (UTC)

  Now updated in quarry:query/107210 (and runs faster). (The other two are updated as well; see 107208 and 107209.) Per a comment there by WaId, it may need further tweaks, as it looks like it includes spurious rows, but I think it provides useful info overall. Mathglot (talk) 19:26, 11 July 2026 (UTC)

  I agree. I think retention is great but making just a few decent contributions is more than most will ever do. —Myceteae‍🍄‍🟫 (talk) 15:28, 10 July 2026 (UTC)

  When this question comes up, I always like to mention myself as a case study! I originally created my account for a WikiEd course in 2018 (during which I rewrote Miltiades the Elder), and while I didn't become a consistently active editor until 2021, I credit the WikiEd course with giving me confidence in my ability to successfully edit. My edits back during WikiEd weren't perfect by any means (I cringe a bit to see myself straight-facedly citing Herodotus in the Miltiades article lol), but I've sharpened my skills since then, and nowadays have 15k edits and a GA to my name. ModernDayTrilobite (talk • contribs) 17:08, 10 July 2026 (UTC)

  I wonder if there are any Wikipedia:Userboxes for former student editors. WhatamIdoing (talk) 17:36, 10 July 2026 (UTC)

  That's a fun idea! —Myceteae‍🍄‍🟫 (talk) 17:38, 10 July 2026 (UTC)

  Google's standalone conversational AI said there wasn't, so here's what I came up with:

Wikitext	userbox	where used
{{User:Clarinetguy097/Userboxes/User former student editor}}	WikiEdThis user is a former student editor.	linked pages

Clarinetguy097 (talk) 01:59, 11 July 2026 (UTC)

Adjusted indentation per WP:THREAD. Mathglot (talk) 02:06, 11 July 2026 (UTC)

• I know of a couple other admins and several other active users who came to Wikipedia through a class project. I won't name them because I don't know how comfortable they'd be with it. There's another category worth mentioning, too: instructor who have become regulars since teaching with Wikipedia. I haven't taught in a while, but incorporated it into a few classes way-back-when. I'd been mostly a wikilurker and jumped into active editing because I wanted to make sure I knew what I was talking about (this was before WikiEdu, with its dashboard and training modules and staff to help with student problems, so I was mostly on my own). I know a bunch of other people in that category, too. But at the end of the day, retention isn't the point of these projects. The point, from a Wikipedia point of view, is to make sure the time students do spend here is beneficial, on the whole, for the project (which doesn't mean problem-free, or even that there aren't problem classes -- with hundreds every semester, there's bound to be a couple, but it's still a much better ratio than the general new editor pool, as pointed out by others above). — Rhododendrites ^talk \\ 17:58, 10 July 2026 (UTC)

• I'm convinced by the arguments in this thread that WikiEd should stay, mostly because teachers will have kids do it anyways. I am also convinced from monitoring the 1346 (hist · log) and 1325 (hist · log) filter logs every day last November/December that this problem is not as under control as some are suggesting it is, despite genuinely very responsive action from WikiEd staff like Brianda. I'd like to see suggestions for how we can better manage student-created content, such as additional restrictions around creating new articles. NicheSports (talk) 21:31, 9 July 2026 (UTC)
• I've been puzzling a bit over the perception of student edits as extremely detrimental versus the statistics contradicting that, and I wonder if some of this doesn't lie in the way typical student editors "mess up" articles versus the way the median new user does the same. Typically when we see someone adding "bad" content to an article we're watching, it's pretty easy to reach out and revert with one of the all-purposes excuses: this is OR, it's unsourced, NOTHOWTO...and so forth. Student work is more focused on one or a few articles, and I suspect that typical problems with it are issues of tone (reading like a persuasive essay) and balance (larding up the article with material that's trivial, or peripheral, or excessive background). So it's too good for a blanket revert, problematic enough to disfigure the article, and addressing the tone and balance issues and re-integrating with existing content requires a significant investment of time. Choess (talk) 04:23, 12 July 2026 (UTC)

  I still think the program should go, but I heavily agree with your final bit about hard to remove bad edits. I honestly feel that there should be a WikiProject for fixing articles with complex issues like this. Ilov3gam3z (talk) 11:54, 12 July 2026 (UTC)

  So the problem isn't obviously bad edits (which would take maybe three seconds to remove); instead, the "problem" is that most student edits have some obvious value, and responding correctly to them takes longer than a few seconds?

  I wonder if there should be something in WP:OWNBEHAVIOR about discouraging other people from contributing because it's all just too much work for us to collaborate. WhatamIdoing (talk) 16:25, 12 July 2026 (UTC)

  The work really adds up when there are multiple courses running at the same time. Clarinetguy097 (talk) 18:03, 12 July 2026 (UTC)

  Yeah. Most of the time the problems span multiple articles with too many edits to revert, (3RR) over large periods of time and in a variety of places in each article. The problems are way bigger then many seem to be making them out to be. Ilov3gam3z (talk) 18:25, 12 July 2026 (UTC)

  Most student editors only edit one or two articles at the most, and unlike other new editors, they're trained on requirements like 3RR.

  Are you perhaps blaming each student for all the other students' edits, instead of just for their own? WhatamIdoing (talk) 19:04, 12 July 2026 (UTC)

  It's definitely a fact that student edits, collectively, can "mess up" things in ways we don't see for most other inexperienced editors. When a class of however many students makes changes to a group of articles, the quantity of edits exceeds what individual, unconnected new editors typically do (by a factor of "however many" students there are in the class). And because student edits tend to manifest just before the end of the semester, they typically all come in at about the same time, whereas the appearance of other new editors is stochastically distributed over time. Taken together, these things make student edits more impactful, all at once, than the typical drip-drip of other newbie edits that we see on our watchlists. It's worth noting, however, that this is true of student edits collectively, but not of student edits individually.

  If the class project is poorly designed, the student edits may also be oriented towards a grading metric that is misaligned with Wikipedia's expectations – something where WikiEd is a good influence. And there's an odd kind of psychology in which many experienced editors feel reluctant to interfere with student edits, in ways they wouldn't for edits by anyone else, because there's something cringey about undoing something that some young person might "need" for a grade. That's something that editors ought to get over, and treat student edits as just as revertable as any others, remembering that we are not supposed to be unpaid teaching assistants. And WikiEd helps there to, by advising instructors never to grade based upon whether or not edits are retained. The problem in this discussion thread is that editors sometimes equate "bad student edits" with "all student edits" with "WikiEd's fault" because the pattern tends to stand out, even if it is not a logical correlation. --Tryptofish (talk) 20:22, 12 July 2026 (UTC)

  Another thing I forgot is that student editors also sometimes dump a huge change to a page, that they copy-pasted from their sandbox. That's another thing WikiEd has been working to discourage. --Tryptofish (talk) 20:47, 12 July 2026 (UTC)

  These editors edit over a longer period of time, sometimes in between other users good edits, at high frequency and in ways that guarantee edit conflicts. Student editors make very complex and systemic problems to articles then what normal, inexperienced newbies create. It can be quite the headache. This is my reasoning. Ilov3gam3z (talk) 20:43, 12 July 2026 (UTC)

  Do you think you could reframe your reasoning to sound less like "How dare they try to make significant contributions, and don't just give up after a day or two of me reverting them, like all the other normal, easily discouraged newcomers"?

  I've dealt with student editors for years. Some of the students (and classes) are better or worse than others. Sometimes I feel like we have to intervene and explain why something isn't working (e.g., in medical articles for the last couple of years, we have seen inappropriate use of gender-neutral language for women's health subjects; anything about eating disorders usually devolves into an advocacy piece). Occasionally, I'll even give up and revert everything a student did (if it's a big end-of-class edit, I suggest waiting until the class is over, because it's less likely to result in the student re-adding the changes the next day).

  But even though there are problems, the articles are usually improved, or the edits show us the need for improvements that we hadn't thought about. It's our official policy that WP:IMPERFECT contributions are acceptable, because if we work together, the articles will eventually be improved. Driving away student editors gives us a superficial "peace for our time", but it ultimately harms the project. WhatamIdoing (talk) 22:15, 12 July 2026 (UTC)

  Your framing of my argument is illogical, wrong and insulting. Did we forget this comment you made at my discussion in village pump on biased student edits:

  When the problem is missing perspectives, then "adding bias the other way" actually gets the article closer to being neutral.

  I am done with this thread. Ilov3gam3z (talk) 22:33, 12 July 2026 (UTC)

  • I'm telling you accurately how your complaint strikes me. You complain that "These editors edit over a longer period of time" – as if that were a bad thing. You complain that "Student editors make very complex" edits compared to untrained newcomers – as if that were a bad thing. You complain that they edit "multiple articles" – even though most don't, and even though editing multiple articles is generally considered a good thing.
  • As we discussed earlier, when the problem in the article is a violation of the WP:YESPOV policy, then adding the missing POVs is how we fix the article. You cannot fix a problem of missing perspectives by continuing to exclude the missing the perspective.

  WhatamIdoing (talk) 23:14, 12 July 2026 (UTC)

Should WikiEd be advertising?

At the related Village Pump discussion, ElKevbo told me about something I hadn't known about before: (and my reply: ). I think that this is something worth discussing, perhaps in more detail, here as well. I learned that WikiEd does some sort of outreach (I don't claim to know the specifics) to instructors over a mailing list. As I said in that other discussion, I think it's fine for WikiEd to get out the message that if an instructor is considering doing a class project here, the instructor needs to be aware of WikiEd and work with them. But I also think that there can be a downside if the message crosses over to something like we are looking to get more class projects here, so consider sending your students our way. WikiEd has a small staff, relative to the numbers of students they have to advise at any given time, and it does nobody any good if they get overextended. I realize that there is a culture at the WMF of "always growing", but I also think that the editing community can potentially give useful feedback about this. So that's why I'm starting this discussion. --Tryptofish (talk) 17:00, 13 July 2026 (UTC)

That discussion is at Wikipedia:Village pump (miscellaneous)#Wiki Education is harming articles related to Native Americans. WhatamIdoing (talk) 17:13, 13 July 2026 (UTC)

Currently, this is being discussed more there, than here, in case anyone here wants to look there. --Tryptofish (talk) 23:15, 14 July 2026 (UTC)

Must. Refrain. From. Raising. This. At. Talk:Timeline of the far future#Wiki Ed still causing strife. Take a breath, Mathglot. Mathglot (talk) 23:36, 14 July 2026 (UTC)

The content of what was posted on that academic mailing list has now been copied to here: .

@Helaine (Wiki Ed): I have some very serious concerns about this.

It comes across as telling instructors who might not have previously been planning to create class assignments on Wikipedia that they ought to come here: With the Wikipedia Assignment, students do the intellectual heavy-lifting: researching, fact-checking, writing, and ultimately publishing to one of the world’s most-viewed sources of information. This fall, swap out a writing assignment for the Wikipedia Assignment and empower your students to use their human critical thinking skills to make a real-world impact on public knowledge. Each term, nonprofit Wiki Education provides free support to hundreds of postsecondary courses across the U.S. and Canada for their projects on Wikipedia — yours can be next!... get started today — no need to attend an information session first. That can be construed as something where the students "do the... heavy-lifting" but that will be less work for an instructor who can now "swap out a writing assignment" that they would have had to read and grade themselves.

It's posed as something where students won't be able to rely on AI: Free support for an assignment AI can’t shortcut. AI can write a term paper in seconds — but you can’t be sure that it verified the facts and created knowledge readers can actually trust. But that actually means that Wikipedia editors might have to patrol for AI slop, instead of the instructor, at a time when AI slop is perceived by the community as a serious and growing problem (a kind of "support" that is not "free" for the editing community).

WikiEd staff are limited in numbers, and it's not at all clear that increasing the quantity of class projects is compatible with the staff providing close supervision for all the students. I think reasonable editors agree that it's good to make instructors in the US and Canada aware that WikiEd exists and is important if they already are considering coming here, but that's not what I think this message will accomplish. There is a real risk of creating hassles for the rest of the editing community, and for that to result in bad will. --Tryptofish (talk) 21:04, 15 July 2026 (UTC)

Ohhhh boy. This is not good. For one, I have no idea why the implication is that it's harder to AI-generate a wikipedia article than a student essay, or why students on Wikipedia are more likely to actually read what they're citing if they're writing for this platform and not for a student paper. So I consider this misleading to the instructors. On the other hand, it's easy to immediately get a list of fellow WikiEdu editors on a course that's gone bad, so I'm not really concerned about this causing undue disruption. It's trivially easy to block students who use AI to generate articles. I will do it for free any day of the week. @Helaine (Wiki Ed), I hope you will make instructors aware that students who use AI on Wikipedia will be blocked from editing, and that unblocks admins are not particularly interested in spending their time helping out students who have been blocked for this reason. Since I am aware that WikiEdu training materials explain that LLM-generated material is not allowed on Wikipedia, I personally would consider that sufficiently clear prior warning that I would block without hesitation in those circumstances. In solidarity, asilvering (talk) 21:44, 15 July 2026 (UTC)

And, for what it's worth: @Helaine (Wiki Ed) and others, if you want to ask for enough to pay for a new staff member in your next WMF grant proposal, you certainly have my endorsement. In solidarity, asilvering (talk) 21:59, 15 July 2026 (UTC)

@Tryptofish, two immediate thoughts:

• I don't know any instructor who ever thought that swapping out a familiar assignment for a new thing (on a completely unfamiliar platform, no less) was ever going to be less work for them. In my experience, it's more typical to hear complaints about even small changes, such as when a new edition of a textbook comes out, because you've been assigning "chapter 8, questions 1, 4–6, and 9" for the last five years, and now you have to find out whether the questions have been changed or re-ordered.
• Wiki Edu scans student contributions for AI use. I believe I've mentioned this before. Please read that, paying attention to sentences such as "Thanks to in-kind support from our friends at Pangram, we began running our participants’ Wikipedia edits, including in their sandboxes, through Pangram nearly in real time". This doesn't "actually means that Wikipedia editors might have to patrol for AI slop, instead of the instructor". (It might mean that we ought to be asking them how we can get the tools here. Imagine if an AI "copyeditor" got a warning message from Special:AbuseFilter before they dumped their LLM edits into an article.)

WhatamIdoing (talk) 22:49, 15 July 2026 (UTC)

On the second point, I'm glad to know that WikiEd scans for that, although other editors who patrol for this might well find it first. As for the first point, I've worked in academia for a long time, and I know that it's dangerous to generalize about that. There are plenty of instructors at some institutions who are underpaid non-tenure track employees who are under a lot of pressure to process a lot of tuition-paying students, where the incentives are to maximize cash flow and minimize friction, and it can be easy to sell them on the idea that they can send students here and it will look good to their supervisors, and there will be less time reading and grading papers. But in any case, none of these points supports the idea that advertising presented this way will be a good thing. --Tryptofish (talk) 22:58, 15 July 2026 (UTC)

Just anecdotally speaking, I feel like I haven't encountered many AI edits from WikiEdu students here in the past few months. It could be that I'm just not finding them, or that my patrolling is targeting older LLM output. But it could also mean that the scanning pipeline is working. Gnomingstuff (talk) 02:31, 16 July 2026 (UTC)

(to be more clear, I do not see the cause-effect relationship between "WikiEdu sends out a normal-sounding marketing email" and "people now have to patrol for AI slop," given that people have to patrol for AI slop already, and most of it does not seem to be coming from students these days.) Gnomingstuff (talk) 02:38, 16 July 2026 (UTC)

That's good news, thanks. (I know you do a lot of work in this area, and I've seen your recent posts about finding it discouraging.) I wouldn't want this kind of outreach from WikiEd to risk making the problem worse, and it would be pretty easy to revise the post to make it less prone to misinterpretation. --Tryptofish (talk) 21:03, 16 July 2026 (UTC)

That's what I'm saying -- I don't see how this is likely to make the problem worse. What is there to misinterpret? This email is marketing to professors who don't want their students using AI. If a professor doesn't care about their students using AI, or wants them to use AI, then this email isn't going to do much for them.

There's also scale: due to the nature of the assignment, each student concentrates edits on one article, maybe two or three. So even if every single student in a class is using AI and somehow gets it through the detection pipeline, they're still taking several months to produce the amount of AI content that the average AI-using editor here churns out in a week.

(re: getting a similar pipeline here, everyone would love that; there's been a lot of work in producing better detection tools and some stuff in the works for proactive detection tools, although I'm not directly involved) Gnomingstuff (talk) 13:32, 17 July 2026 (UTC)

An important data point that I don't have yet is how long these mailing list postings have been going on for, and what wording has been used in the past. If, hypothetically, this was the first time that the post referred to AI in the way that it does, then that would negate your experience, because next semester will be the first time that class projects show up here in response to that mailing list post. Maybe there will be more classes from careless instructors who don't sufficiently discourage AI, or maybe not, but we don't know yet. I've posted here about the way an advert was constructed, and you've replied to me in terms of what you have experienced before that advert was ever posted (and all of this is happening against the backdrop of AI use by students increasing rapidly in society as a whole).

Another issue is something you allude to when you say "somehow gets it through the detection pipeline". I don't know how well or how promptly the screening used by WikiEd works, although I'm certainly hopeful.

Finally, what you said about scale is not accurate for class projects. Most student edits come in just before the end of the semester, usually within a period of a few days, and there can be many students in a class, so the issue here is the sum of all those student edits, not just the edits of one student. That pattern has historically been something that has prompted editors to get upset about class projects. (I still remember the thread on this noticeboard from several years ago, about a class that showed up with hundreds of students, all editing badly.) --Tryptofish (talk) 21:39, 17 July 2026 (UTC)

I find Wiki Ed recruiting messages in the POD mailing list archives going back to at least 2016. The contents of the messages have changed, of course, and I haven't tracked how they have changed over time except that it appears that explicit mention of AI only began this June. And I haven't looked at the materials that are referenced in any of the messages (e.g., many include a link a testimonial from a professor about their experience, most/all include links to Wiki Ed training materials) to see if AI is mentioned in them. ElKevbo (talk) 22:22, 17 July 2026 (UTC)

the issue here is the sum of all those student edits, not just the edits of one student

Again, that's what I said. Suppose you have a really big class, 100 students, and every single one of them both used AI and were able to get it through the pipeline. That seems like a lot, but there are some people with open WP:AINB threads who have hundreds of AI edits on their own.

As far as the detection process they just wrote a post on that, I started a new thread for it below. Gnomingstuff (talk) 23:43, 17 July 2026 (UTC)

@WhatamIdoing, a lot of instructors these days, faced with tides of AI-generated slop essays, are looking for alternatives. This email presents an alternative, but, in my opinion, does so misleadingly. In solidarity, asilvering (talk) 00:20, 16 July 2026 (UTC)

Note: I've posted about this to editors who are concerned about AI editing: . --Tryptofish (talk) 21:10, 15 July 2026 (UTC)

I was not aware of this particular type of listserv post, but I guess I'm surprised that others are surprised by this. We've known that Wiki Ed has various growth and outreach initiatives. I hadn't given much thought to their form or content but I can't say this is unexpected. —Myceteae‍🍄‍🟫 (talk) 20:53, 16 July 2026 (UTC)

Fair enough to have a conversation about wording and expectations, but as far as "should WikiEdu be allowed to tell professors it exists and that teaching with Wikipedia is a good idea" ... Where do we imagine funding comes from for a secret organization that isn't allowed to tell anyone proactively that it exists, and serves only a reactive support role? Because it's certainly not the WMF or any funder I'm aware of. We can say "oh, well, the WMF should fund that, and five more staff while they're at it" but that doesn't translate to reality. — Rhododendrites ^talk \\ 23:07, 15 July 2026 (UTC)

It occurs to me that I should point out that, when I titled this thread the way I did, I was making a word-play on the title of the main section thread, and that maybe that intended irony is backfiring. In case it's not sufficiently clear, I don't have a problem with WikiEd making themselves known, and I'm not trying to interfere with their funding. What I want is for them to be careful in how they word their outreach, so as not to attract unwanted results. --Tryptofish (talk) 21:03, 16 July 2026 (UTC)

The WMF's general rule, going back many years, is for organizations to get less than 50% of their funding from the WMF, especially if the organization isn't new and isn't in a low-income country. This is a normal guideline for grant funders (a common goal is to be between 5% and 50% of a program or organization – big enough to make a difference, but small enough that if you cancel your grant, it won't fold overnight), so I don't think that having a bunch of random people on the internet saying "oh, well, the WMF should fund that" is going to change that goal. (I haven't checked, but I suspect that Wiki Edu gets significantly less than 50% of their funding from the WMF, but realistically, 50% is going to be a hard ceiling.) WhatamIdoing (talk) 01:11, 16 July 2026 (UTC)

Guidance on current events articles

Should we add guidance to WP:ASSIGN discouraging students from editing articles with the {{Current}} tag for class assignments? Today I saw that a student has selected MV Hondius hantavirus outbreak for a Wiki Ed class assignment. I went to the student's user talk page and saw that the Wiki Ed rep for the class had already left them a suggestion about choosing a different topic, mostly related to WP:MEDRS concerns, which are covered at WP:ASSIGN#Editing medicine and health topics.

I chimed in to add that there are a whole host of challenges to editing articles about high profile current events. These articles strike me as especially poorly suited for class assignments. The frequent edits, constantly changing facts on the ground, and fast-moving talk page discussion make it nearly impossible to do the kind of content assessment and well-planned edits that I understand to be a part of the best assignments. —Myceteae‍🍄‍🟫 (talk) 21:32, 13 May 2026 (UTC)

Good idea, thanks. Done: . --Tryptofish (talk) 21:58, 13 May 2026 (UTC)

Thanks! Perhaps it could be fleshed out a bit and separated as a brief standalone paragraph to distinguish it from contentious topics. Many breaking news stories will touch one or more contentious topics but many don't and there is a unique set of issues at play even when the underlying subject matter is not obviously "controversial" and doesn't fall into one of our CTOPs. An edit that was perfectly placed this morning can be problematic by lunchtime. —Myceteae‍🍄‍🟫 (talk) 22:49, 13 May 2026 (UTC)

I worry that ASSIGN is getting too long, so I'm kind of thinking that lots of stuff should be shortened, instead of making anything longer. This could also be discussed at WT:ASSIGN. --Tryptofish (talk) 22:52, 13 May 2026 (UTC)

Having just become aware of it today, I tend to agree it's too long but something also feels missing. Looking at the Wiki Ed expert's user talk page today, there were a couple similar cases raised, although these articles had other issues like CTOP so it wasn't an issue of not being addressed by the guidance. I may ponder it and propose something more concrete. —Myceteae‍🍄‍🟫 (talk) 00:10, 14 May 2026 (UTC)

I wouldn't worry about the length too much. By college standards it's light reading. Clarinetguy097 (talk) 16:09, 15 May 2026 (UTC)

I see your change as a further improvement. This separates it from CTOPs and gives a little more information without substantially adding to the length. —Myceteae‍🍄‍🟫 (talk) 20:34, 16 May 2026 (UTC)

I like that edit too, and I appreciate the observation about not being long by the standards of college reading assignments. --Tryptofish (talk) 20:38, 16 May 2026 (UTC)

Could a flowchart be a helpful solution here? e.g., "Does the article exist already?" "Do I have a personal or professional relationship with the article subject?" "Is the article about a medical topic?" "Does the article relate to US politics after 1992?" (or does this already exist?) Wracking ^talk! 23:21, 15 July 2026 (UTC)

I don't know if there is one, but if not, I think it's an excellent suggestion, @Wracking. Schazjmd (talk) 23:25, 15 July 2026 (UTC)

Help:Your first article covers a lot of those points, in a step-by-step manner, though not all. Mathglot (talk) 23:30, 15 July 2026 (UTC)

Agree, and just wanted to strengthen the argument with another reason to avoid {{current event}}s, namely, the very high likelihood of experiencing edit conflicts. I experienced numerous ec's editing the MV Hondius article. At times I had to alter my normal editing procedure and rather than edit a couple of sentences in two different sections sharing the same re-usable ref, say, I would do it in two very quick, bite-size edits because trying to do it in one would never stick. That's maybe okay for a senior editor (and still no picnic) but is a mine field for the average editor. New users are particularly ill-suited to deal with edit conflicts much less strings of them that might look like they are doing something wrong when they aren't. Ignoring the ec-resolution instructions leaves clobbered edits behind for other users to find (if they ever do) and clean up afterward. So, yeah, avoiding steering students to current event articles is a great idea, for multiple reasons. Mathglot (talk) 19:20, 20 May 2026 (UTC)

Oh, I guess you mentioned that already, as 'frequent edits'; I searched for 'conflict' but should have read the whole thing. Anyway, I agree, and can only underline the point. Mathglot (talk)

I wasn't thinking of edit conflicts specifically, but you're right, these are a special case of 'frequent edits' that can cause particular difficulty. —Myceteae‍🍄‍🟫 (talk) 20:33, 20 May 2026 (UTC)

Is the other reason that students will be competing with editors to add content? Students need to contribute an adequate amount to get evaluated, but there'll be a lot of other people working too (if the topic is popular enough, people might even be working around the clock). Clarinetguy097 (talk) 03:23, 21 May 2026 (UTC)

Yes, that was part of my thinking, at least sort of. Obviously every assignment is different and the larger Wikipedia community can't control and isn't responsible for student course outcomes. But most assignments, as I understand it, involve selecting a topic, evaluating the current state of on-wiki coverage, identifying meaningful gaps that can be filled, and making substantive edits to improve the content. A well designed assignment that is well executed probably involves a lot of drafting and review before publishing an edit. Pretty much every step of this will be thwarted if students select a breaking news topic. The current state of these articles is unstable and any assessment of coverage will likely be irrelevant within a few hours. If editors take their time to research and draft contributions, both the article and real-world facts will very likely have changed by the time they go to publish. Even high quality additions often don't survive long or are radically transformed because the facts on the ground change or article content is reorganized. These articles just don't seem conducive to WikiEd assignments given the constant change. —Myceteae‍🍄‍🟫 (talk) 04:02, 21 May 2026 (UTC)

Given that the sun never sets on the Wikipedia Empire, people are *definitely* editing current events articles around the clock. Mathglot (talk) 00:39, 7 July 2026 (UTC)

Education financing in <CountryName>

Adapted from a similar post at WT:SPI.

I have noticed a recent flurry of articles or drafts named, 'Education financing in <CountryName>'. Could this be students in an unregistered class? Or is it likely to be a couple dozen users honing in on the same topic for some reason? (Note: This navbox is just a convenient way to list them all here; I am not advocating for creation or use of a navbox like this in articles.)

I know that the 'unregistered class' question is a pretty frequent one at this board, and it would be handy to maybe have a {{FAQ}} that mentioned it, as I forget your usual procedures about how to look into candidate unregistered classes, and what, if anything, to do if it appears to be one. Are there any tools specifically for this question, or how do you proceed, usually? Thanks, Mathglot (talk) 18:59, 28 May 2026 (UTC)

Hi @Mathglot:! These can be classes or editathons; the usual procedure is to try to find out any information you can about where they're coming from. If they are a class, Wiki Education will try to reach out to the instructor if they're in the U.S. or Canada higher education system; if not US/Canada, I can try to connect them with someone more relevant for their context. For an editathon, they can usually tell you who organized it, and that person is a better point of contact. --LiAnna (Wiki Ed) (talk) 21:58, 28 May 2026 (UTC)

Did anyone figure out if this is a class or not? It seems that some of the articles have characteristics of essays and feature long disgressions on the countries' economies, so the authors probably need help. Clarinetguy097 (talk) 15:36, 31 May 2026 (UTC)

Spot-checking just a small handful of them, I had concerns about AI-generated content. User:Gnomingstuff has been involved with that and might have comments, or maybe we should just raise this or link it from the WP:AI noticeboard. Mathglot (talk) 15:50, 31 May 2026 (UTC)

Hi - the AI noticeboard is the best place for stuff like this in general, although if it’s a college class specifically the professors may already be looped in.

I don’t have time currently to look over these, apologies, but maybe people can help there. Gnomingstuff (talk) 16:32, 31 May 2026 (UTC)

I think the AI is mixed. Didn't see any strong signs in Singapore, but South Africa is blatantly llm. CMD (talk) 16:12, 7 July 2026 (UTC)

Somebody at the AfD has said it was a university assignment. Clarinetguy097 (talk) 14:17, 6 July 2026 (UTC)

Setting {{subst:DNAU|30}} here, as this section is {{Excerpt}}ed in the ongoing Afd. Mathglot (talk) 23:42, 6 July 2026 (UTC)

Class outside WikiEdu

Just stumbled across this class due to this unfortunate edit. Just wanted to provide a heads-up here in case someone can support the class. -- Avocado (talk) 21:29, 10 June 2026 (UTC)

ping @Earthdirt:! --LiAnna (Wiki Ed) (talk) 22:25, 10 June 2026 (UTC)

Just noting that, per the linked page, it appears this teacher has been running this project annually since 2012. I'm curious whether the instructor has been aware of Wiki Ed previously. —Myceteae‍🌈 (talk) 22:37, 10 June 2026 (UTC)

I did post a note on the instructor's talk page, too, in part out of curiosity about that very question -- Avocado (talk) 22:42, 10 June 2026 (UTC)

Hi @Avocado, @Myceteae, & @LiAnna (Wiki Ed). I have never heard of this noticeboard/project before. What is it exactly that you would like me to do for future projects regarding it? It's quite unclear what this is, perhaps just notice that my project exists for you? I'm a huge fan and longtime editor of Wikipedia and the whole foundation and am committed to having my students contribute scientific images, many of my student contributions over the years have added significant value to articles across Wikimedia projects. Certainly not all will be good nor will all take/use editorial review to make appropriate images (though many of those are high quality and may go on to be used outside of Wikipedia on external sites/projects). In the past 14 years none of my project's students have been involved in vandalism or other serious issues (other than a 2 cases of plagiarism/AI usage sadly including one this year). Please know that I review all submissions and edits at the end of the project (this year June 16) and will remove/edit any diagrams I do not think are appropriate if other editors have not already. Thanks for your work on behalf of wikipedia and education! Cheers, Earthdirt (talk) 15:37, 11 June 2026 (UTC)

Just noting here that the discussion has continued on the user's talk page at User talk:Earthdirt#Student assignments. I have responded there. —Myceteae‍🌈 (talk) 23:17, 11 June 2026 (UTC)

Just a status update: they responded there on 12 June, and not since. Their last Wikipedia edit was a few minutes later. Mathglot (talk) 19:09, 9 July 2026 (UTC)

I don't feel this needs further follow up or monitoring, although it was appropriate to raise here. —Myceteae‍🍄‍🟫 (talk) 20:59, 9 July 2026 (UTC)

Discussion at Wikipedia:Village pump (miscellaneous) § Wiki Education is harming articles related to Native Americans

 You are invited to join the discussion at Wikipedia:Village pump (miscellaneous) § Wiki Education is harming articles related to Native Americans. —Myceteae‍🍄‍🟫 (talk) 21:31, 9 July 2026 (UTC)

Brainstorming a FAQ

I think it's pretty clear there are some questions that come up frequently, and I think we could all benefit by having a FAQ (in some cases, with archive links to repeated discussions). To that end, I want to start a brainstorming discussion, which could just be a list or grab-bag of what the most common issues are that come up, and then winnow it down later to a reasonably sized {{FAQ}} we can incorporate in the header.

So, please put on your brainstorming cap and add your thoughts about frequent questions below. Can be just a bullet with a question, with or without an "answer" or links to archives, or whatever you like. Please formulate in terms of a question (unnumbered, for now) and add any remarks indented just below it. I'll start:

• What should we do about an unofficial/undeclared class I found?

  Note: this Cirrus search finds eight discussions in the archives (still working on this search).

• A lot of students make poor edits. Wouldn't it be better for Wikipedia if we ended/restricted the program?

  Links: § Should WikiEd be terminated?, [more later...]

• Are certain topics unsuitable (or, 'off-limits') for students?

  perhaps controversial topics, current events, high-traffic articles, featured articles, articles under dispute (active Rfcs, edit-warring, WP:AE, etc.), special requirements (WP:MEDRS, WP:BLP...). Maybe just the top 3, for starters?

• A student made a poor edit. Should I revert it?

  Suggest: Treat a student like any other editor.

• Where should I raise concerns about problem edits? Article TP? Student UTP? Instructor? Wiki Ed expert?

  tbd

• Our class is in OtherCountry. Can we join Wiki Ed?

  There seems to be a standard boxed reply in the header already; perhaps move that to the FAQ, along with a link to affiliated programs; I know India has one; forget where the page is.

Please add your thoughts about questions that ought to be part of a FAQ, or comments on the ones above. Let's focus on questions for now; we can argue about answers later. Thanks, Mathglot (talk) 22:47, 10 July 2026 (UTC)

This is a great idea, thank you for starting it!! I’d add that people often seem unaware of what exact training materials instructors / students see, so that would be good to provide, and maybe also an answer for “how can I see edits make through wikied?” giving a link to their dashboard ecosystem. ~ le 🌸 valyn (talk) 03:43, 11 July 2026 (UTC)

Or, "where can I find articles that are part of current and/or past WikiEd projects"? Clarinetguy097 (talk) 16:53, 12 July 2026 (UTC)

That's a technical question, but not a very frequent one, I don't think. Here's how:

Articles for past Wiki Ed campaigns

To find articles for a past Wiki Ed campaign you can do this:

• start at the Wiki Ed Campaigns page
• in the search box, search for the year you are interested in, e.g. '2025' to narrow the list
  • e.g. search 2025 to get to the list of 2025 campaigns
• Pick a campaign, to display the overview page
  • e.g., click 'Fall 2025' to get the Fall 2025 overview page.
• Select 'Articles' in the top links bar to get the first page of 25 out of about ~6,275 articles. (unsorted)
• the top five artiicles shown on page 1 of results are: AI anthropomorphism, Nutritional epigenetics, Chronic traumatic encephalopathy in sports, Ling Cage, and Human-AI interaction.

Note: clicking col. header 'title' sorts the results on that page only, not the list.

Thanks, Mathglot (talk) 19:14, 12 July 2026 (UTC)

Maybe:

Aren't all classes required to work with Wiki Edu Foundation?

Wiki Edu only supports classes in the US and Canada. Sometimes other affiliates support classes in other countries. There is no requirement that students be supported by any organization.

I'm an instructor. Is there a way for me to talk to other instructors about what this is really like?

See [link to possible mentors?].

Do any students keep editing after the class?

Yes, some admins and other long-time editors began editing as students, and others began editing as professors who wanted to understand how to edit Wikipedia so they could encourage their students to contribute. One estimate is that about 5% of editors involved in a class might continue editing after a class, which is a higher retention rate than non-student editors.

The last estimate comes from the recent WP:RAQ request; it should be replaced with stronger/cleaner numbers (perhaps we should standardize that on one year after the first edit, to make the comparison easier). WhatamIdoing (talk) 20:34, 15 July 2026 (UTC)

Point by point:

• 1) agreed.
• 2) A bit confused; in your bracket-thetical, did you mean, 'other possible instructors', or did you intend to say 'mentors'? I think having a way to contact other instructors would be great! How 'bout, Wikipedia:Education Faculty lounge? On the downside, although I interact a fair bit with Wiki Ed experts and students, my interaction with instructors has been minimal, and when I have made overtures to them, I rarely hear back (with a couple of exceptions who have been active at WP well outside their courses). So I suspect nobody would hang out at the Lounge, although experienced instructors would surely be a unique and important resource for other instructors just starting out with Wiki Ed. Adding Piotrus.
• 3) Could we change the Q wording to a quantifying question rather than yes-no, e.g., 'How many...', or, 'What percentage of...' ? The RAQ numbers are from 2023; need input from Sage to get rerun the query and get better quality numbers based on the latest student list. (Ultimately, we should run it periodically—monthly perhaps—and store summary values in .tab files at Commons (e.g., sample .tab file) so we can chart them (e.g., sample chart file). I could see a summary chart with three lines: number of wiki students, number of edits X-interval post course-end, and the ratio of the two. (Running the query several times with different values for X would provide a lot more data very cheaply; e.g., X=3 months, 6, 12, 24, 36, 60, 120). The first two lines of each triplet measure growth, and to an extent, the last line would be a measure of how effective our student retention efforts are.)

Thanks, Mathglot (talk) 01:50, 18 July 2026 (UTC)

1. Okay
2. I doubt that an on-wiki page would be functional; e-mail is more likely to work. I was thinking that we might be able to come up with a list of long-time editors such as @Rhododendrites, @OhanaUnited, or Piotrus who would be willing to answer questions. Wiki Edu probably has their own list, but that's not helpful to instructors (such as Piotrus) who aren't in the US or Canada.
3. I hope and expect you to change the wording to anything that seems good to you. I suggest, though, that an annual update on numbers would be more pointful than a monthly one.

WhatamIdoing (talk) 02:05, 18 July 2026 (UTC)

In hindsight, I wondered what would happen if I didn't declare my course. Would someone drag me (or my students) to this noticeboard for going stealth? I have definitely thought about not declaring, knowing that there are experienced editors who will be more harsh towards student editors knowing that they are students or because of the experienced editors' previous bad experience with an education project. OhanaUnited^{Talk page} 22:19, 18 July 2026 (UTC)

That's interesting, and in a way, surprising to me because I guess I assumed the opposite. While I hope and believe that I would react in the same way to a student editor and any new editor given the same behavior on their part, I generally feel more worried about the behavior of a non-student editor whose path I happen to cross than a student editor, because in the former case I don't know if a single experienced editor other than me is monitoring them (if I even am myself), and therefore what damage the newbie might be doing, either in good faith or with malice.

Whereas with a student editor, at a minimum there's no ill intent (that must be vanishingly rare), plus there are training modules to be done (and their completion status marked), peer reviews from other students, instructors (off-line mostly, but another set of eyes nevertheless), and not least, their Wiki Ed expert(s) and the backing of the entire Wiki Ed organization and ENB, should an unusual situation pop up.

That all leaves me with a sense that a student may be just as new, but they have plenty of help and resources to fall back on, and a community actively shepherding them on board as a new Wikipedia editor from multiple angles. I, for one, won't be more harsh toward a student editor, just far less worried that they may be a young bull in a china shop running around unwatched and untethered, than a non-student might be. I can't speak for others, but if I were going to bring you to this board for going stealth, it would be only to chide you for depriving your students of the benefits and additional support they (and you, and Wikipedia) would receive by being part of it. But I suppose this doesn't help you any, because now it seems like you're damned if you do, and damned if you don't. Mathglot (talk) 22:54, 18 July 2026 (UTC)

This is a very thought-provoking question, so I'll say what my thoughts on it are. OhanaUnited is an editor who has been around a long time and has good social capital here, so whether he did or did not declare would not lead me to worry about somehow going rogue with a class project; I would tend to trust that things would be OK in either case. That would be different if the instructor were someone with no previous editing experience. Whether someone would drag the instructor or the students to a noticeboard would be a function of whether the student edits were perceived as creating problems. Of course, there are some editors who will rush to a noticeboard based on The Rules^TM, but more clueful editors would hopefully not do that. I could imagine a situation where an editor would report here in good faith, and say this appears to be a class that is doing well, but it might be nice if they could connect with WikiEd, and I hope that would not be seen as confrontational.

Now I'll confess some things. When I see a new class project show up on my watchlist, my initial, lizard-brain, reaction is typically one of something like not this again, this is going to be a problem. By the time my higher cortical functions kick in, I feel a little silly about having had that reaction, and I start to look to see what the quality of edits looks like. I'm guessing that other experienced editors have reactions somewhat like that too, so I figured I would throw that out there. A personal concern of mine is when I see student editors working on a CTOP (especially GMOs, because that's something I watch closely), and I've gone to the WikiEd person for that class when that happens, and they are reliably helpful when I do. I know they screen for CTOP editing, but it also looks to me like some student editors slip past that screen, which is a reason why I've been expressing concern in other recent discussions, that we shouldn't be recruiting more classes than the staff can handle.

So how does that feature into how I react to new editors who might be student editors from an undeclared class? It's pretty easy to notice when multiple new editors all show up at once in a topic area on my watchlist. If they self-identify as part of a class project, then that's that. But if they don't, I'm going to wonder if these are sockpuppets (or meatpuppets). And I've noticed noticeboard posts from other editors about the same concern, when it turns out to be either students or an editathon. So one good reason to declare a class is to preclude accusations of socking, which risk being unpleasant for the students. I can't speak to the scenario OhanaUnited describes, about experienced editors who are harsher to declared student editors because of previous bad experiences. It probably happens, largely because student edits show up all at once, instead of spread out over time. But I also see experienced editors who are reluctant to revert students, when they would have no reluctance to revert someone else. So I doubt that there is really enough of an upside to going under the radar, to be a good enough reason to not declare. --Tryptofish (talk) 20:10, 19 July 2026 (UTC)

About your initial, lizard-brain, reaction is typically one of something like not this again: My initial reaction is usually to silently wish them lots of luck, because the article they're looking at usually needs enough work that semi-competent attempts will probably improve it.

In a few cases, my initial reaction is to make a mental note to check back after they're gone, with an expectation that most of it will end up getting reverted (though I've been pleasantly surprised before). I do make an exception for articles on women's health: I have occasionally left a note reminding new editors [students and non-students alike] that being gender neutral is not an excuse for putting innumerate stupidity in an article. "80% of women" is not the same as "80% of people", and the two are not interchangeable when the subject is menstrual cramps, since 50% of people don't have the anatomy necessary to have them. WhatamIdoing (talk) 20:41, 19 July 2026 (UTC)

The first draft of the FAQ is now available. Linked from the header of WP:ENB using template {{FAQ}}, it might look like this:

It has many gaps, and is still a work in progress. Please update the FAQ directly as needed, and discuss here. Ian (Wiki Ed) and Brianda (Wiki Ed), particularly interested in your feedback, as you are the ones seeing and fielding most user questions. Please help: what frequent questions are we missing here, that ought to be listed? Thanks, Mathglot (talk) 09:05, 18 July 2026 (UTC)

We need to elaborate on Q7 - a very important question we need to answer properly, and it also needs to address the issue of what is the instructor responsible for (up to and including some reviewers POV that the instructors should ensure each student article is at Good Article level before they let the students publish it, and that educational projects are a form of PAID initiatives that need to be held to above-average standards - i.e. the instructors are getting paid by the universities, so they should not burden the unpaid volunteers with below-GA content...). As for students, I'd also add something related to "my article was rejected/deleted, what do I do?", which partially can be answered "check the feedback in AfC draft/your talk page, but it also needs to cover the unfortunately common "what do do if there is no feedback, or if it is very generic". _{Piotr Konieczny aka Prokonsul Piotrus| reply here} 08:17, 20 July 2026 (UTC)

(Note: I rearranged some of the questions yesterday. Q7 at the moment is about shutting down Wiki Edu because new students make "poor" edits.) WhatamIdoing (talk) 21:14, 20 July 2026 (UTC)

Yes, I meant current Q7 ("A lot of students make poor edits. Wouldn't it be better for Wikipedia if we ended/restricted the program?"), although since I am am not involved with WikiEdu, I read the "program" more broadly as WP:SUP (which predates WE and has a wider, global scope). Side note: WikiEd is an NGO separate from WMF AFAIK, so "we" cannot close it even if we wanted to :P _{Piotr Konieczny aka Prokonsul Piotrus| reply here} 08:06, 21 July 2026 (UTC)

Reactivation campaign

I'm curious, has anyone tried doing some kind of reactivation campaign for former WikiEdu student participants? I'm imagining this as reaching out on the talk pages of inactive former student editors, especially students who made decent contributions, to encourage them to re-engage as volunteers and offer a pathway back in. For example, if a person edited with WikiEdu 2-5 years ago, they may have graduated and have some time for a hobby that builds on their academic skills. Certainly some of them used university email addresses that have expired, so they wouldn't receive email notifications, but even if a subset could be contacted and were up for re-engaging and becoming longer-term volunteer editors, that could be helpful and nice. Dreamyshade (talk) 15:54, 12 July 2026 (UTC)

Sounds like a good idea. I'm not aware of anyone having tried that recently, if at all. In solidarity, asilvering (talk) 17:07, 12 July 2026 (UTC)

Dreamyshade, see Wikipedia:Education noticeboard/Archive 23#Student editor retention, or re-welcoming. Maybe time to revisit this? See also WP:RAQ#Wiki Ed student retention (redux); a related query might find the inactive ones (on the order of 100k, I think.) Mathglot (talk) 19:54, 12 July 2026 (UTC)

Note in that previous conversation, LiAnna commented that based on their experience with (at that time) over 114,000 students, very few remain as editors, and so they feel they gain more benefit from trying to retain good instructors. isaacl (talk) 22:38, 15 July 2026 (UTC)

Based on my estimates, out of a random sample of 100K new accounts, we only get one editor who will eventually make >100K edits, and we already know enough former-student-current-editors to have exceeded that usual result. WhatamIdoing (talk) 23:05, 15 July 2026 (UTC)

Depends how you interpret "remain as editors"; 100k is a pretty high bar. By your own numbers, 100 remain long enough to make 1,000 edits, so that's a fair indication of longevity, and roughly equates to your output, at least numerically. Counting the other 900 that made it to 100 edits, that's nearly twice your output, and if you believe in the wisdom of the crowd, possibly more valuable, too 😬 (or at least, more varied). Mathglot (talk) 21:55, 20 July 2026 (UTC)

Definitely more varied, and probably more valuable, especially if you're counting visible contributions to the mainspace. (I have a hypothesis that substantial mainspace contributions decline as edit volume rises – the higher the edit count, the more likely that we have stopped adding sections and paragraphs and start tweaking wikitext via AWB and other tools.)

But someone with "only" a thousand edits in a lifetime – someone "only" at the 99th percentile of contributors – will never be able to contribute to policies and guidelines in the same way as someone with even 10K, much less your 100K. We need more thousand-edit editors, but eventually we need a few hundred-thousand-edit editors, too. WhatamIdoing (talk) 22:56, 20 July 2026 (UTC)

I'd suggest doing it sooner than 2–5 years, because many of the accounts will be associated with student e-mail addresses, which will probably disappear at graduation. WhatamIdoing (talk) 18:21, 13 July 2026 (UTC)

Here's a possible approach, but it depends on the assumption that students have access to a non-uni email at end-of-course date. I don't have a ton of contact with generation-high-school, but the few I know all have their own email since tween or teen years; is this generally the case? Let's assume it is true for the majority, and that they keep using it side-by-side with their uni email at college, or at least, they either recover it after graduation or create a new email afterward; is that also a good assumption? If so, then here's a thought.

MW/Outreach has a suggested student survey, and I believe it has been used in the past by Wiki Ed for surveying students, but I am not aware to what extent, or if it is still in use. (Helaine (Wiki Ed), is this accurate?) Suppose we added a new section to the post-course student outreach survey about keeping in touch after their class? In particular, we could explain how to use Preferences to change the e-mail registered on their Wikipedia account, so they could reset it from their uni email (which might disappear) to their former (or new, post-uni) email, along with how to set their preferences to notify them by email when someone writes on their Talk page, explaining that we may contact them again in X years, have a great graduation, enjoy grad school or whatever, and then go quiet. (We might also have to find a way to quiesce irrelevant nonsense from appearing on their Talk page that might cause them annoyance sufficient to cut the email link.) Then, when the time is right, we send them a How's-it-going email, and take it from there. Thoughts? Mathglot (talk) 00:16, 15 July 2026 (UTC)

I don't think I'd want to encourage students quite so strongly to edit under an account directly connected to their WikiEd account -- accounts that participate in WikiEd become permanently much more doxxable than the baseline expectation of anonymity on Wikipedia -- but I definitely think it would be worth sending a follow-up invitation a year or two later to reach the ones who may still have that email connected. It would definitely hurt our metrics-tracking, but in such an email I'd also want to point out the option of making a fresh account for privacy reasons.

But in general, I absolutely think this is the sort of thing worth trying -- WikiEd students strike me as dramatically more 'recruitable' than the general population baseline, and I think it's very valuable to focus recruitment efforts in areas likely to find fertile ground. ~ le 🌸 valyn (talk) 22:36, 15 July 2026 (UTC)

I'm not sure I agree that students are a dramatically more likely population to choose a pastime that involves researching source material to write largely text-based articles. (To create short videos that promote Wikipedia, perhaps.) isaacl (talk) 22:43, 15 July 2026 (UTC)

My thinking was that people who have already made it through several of Wikipedia's 'getting started' hurdles -- such as learning the interface and a few cultural norms -- are 'closer' than those who, e.g., do not yet know that they are allowed to edit Wikipedia. ~ le 🌸 valyn (talk) 22:55, 15 July 2026 (UTC)

Regarding timing, I previously suggested an immediate email at the end of the course, and a follow up a month later. I think the longer the period of time of inactivity, the less likely the user will return to editing, so any prompts should occur early. (In a later comment in that thread, I made a suggestion regarding the email's content.) isaacl (talk) 22:38, 15 July 2026 (UTC)

University students are busy with many crucial things going on in their life, from making it through college, separating from parents, exploring romantic attachments and sexuality, experimenting with their identity, and generally becoming an adult, to figuring out what's next. I'm not sure an early contact request to edit Wikipedia without a very strong, counterbalancing carrot like a course grade would gather much response. But then, that's what mockups and trial programs are for. Mathglot (talk) 22:49, 15 July 2026 (UTC)

Yes, it could be particularly informative to send messages to one cohort who have just finished a course, and, at the same time, another cohort whose course was more than a year ago. And I see little reason not to try both. ~ le 🌸 valyn (talk) 22:53, 15 July 2026 (UTC)

I agree. WhatamIdoing (talk) 23:06, 15 July 2026 (UTC)

I agree that students are busy and are pretty unlikely to remain editors because an email asked them to. But I think that likelihood gets even smaller years later, if that editor can even be reached at that time. Plus it feels a bit spammy to be re-initiating contact years later. isaacl (talk) 22:57, 15 July 2026 (UTC)

Agree with LEvalyn. Above, I suggested both as well, which would reduce the spamminess quotient when the later email arrived. I don't think my university started dunning me for alumni contributions until after a decent interval had passed, and it didn't seem spammy to me at all when they finally did. If we send former students one message early on, I don't see why they would interpret a later contact as spammy, especially as they would certainly recall editing Wikipedia for their course. It's nothing like a cold call, or out-of-the-blue spam. Sure, some ex-students won't welcome it, just like some alumni never cough up a contribution. But many do, and I think it's a worthwhile intiative to try. Mathglot (talk) 23:55, 15 July 2026 (UTC)

Follow-up WikiEdu post on AI and student edits

Balancing out the past few topics with some positivity, WikiEdu just published a follow-up post to their last post. Some excerpts:

• Fewer student posts are getting flagged as AI since they rolled out their Pangram integration in 2025; it's now down to fall 2024 levels. (~10%, down from ~15%). So unless Pangram is returning false negatives more negative (not aware of anything suggesting that it is) it does seem like there's been an improvement.
• If a student post gets flagged as AI, which usually happens at the sandbox/draft stage, the follow-up is this: We then ask the program participant who added the text to confirm the verifiability of each claim, quoting the paragraph they’ve summarized and cited. If they’re able to do so, we determine the text is allowable on Wikipedia; if they are not able to do so, we ask they not move it live again.. This is somewhat more lenient than our guidelines but having done line-by-line fact checking before, "each claim" adds up to a lot of claims, and the last post suggested that articles flagged as AI sometimes fail verification on almost everything.
• Along those lines this seems very promising: Two larger changes we’re rolling out are a new exercise on verifiability and a rethinking of the peer review process to center verifiability. In the new exercise, participants will select an article that included text added to Wikipedia by a past participant flagged by Pangram. They’ll read the source it’s cited to, and attempt to verify the information. Our goal is that by explicitly front-loading the idea that generative AI chatbots often produce unverifiable information, well-meaning participants will understand the drawbacks and avoid using them.

Gnomingstuff (talk) 23:31, 17 July 2026 (UTC)

That was interesting reading, thanks. Two thoughts: 10% of student edits is still a lot of edits, although I agree with you that the overall result is encouraging. Also, a lot of the focus seems to be on verifiability. This obviously makes sense in terms of relating the process to our policies, but there are "tells" of AI-generated text that are unrelated to that, so I would be curious as to how that might affect the reported data. --Tryptofish (talk) 23:47, 17 July 2026 (UTC)

I assume Pangram relies on many of these "AI tells" to flag content as probably AI-generated. The tone, style, and formatting issues are also problems on Wikipedia, but if I had to pick one thing for WikiEd and instructors to focus on most, I'm glad it's verifiability. —Myceteae‍🍄‍🟫 (talk) 00:10, 18 July 2026 (UTC)

Just wondering if Pangram is free to use? I've been using ZeroGPT/GPTZero to check the work of my students (who write for Wikipedia too). My courses are not WikiEdu supported, unfortunately (outside using Outreach Dashboard). _{Piotr Konieczny aka Prokonsul Piotrus| reply here} 08:10, 20 July 2026 (UTC)

https://www.pangram.com/pricing —Myceteae‍🍄‍🟫 (talk) 17:10, 20 July 2026 (UTC)

4 free scans per day, so not useful during heavy-work load periods unless someone has 4 students in class (in my classes I have 50+...). Oh well. Another WikiEdu mostly useless recommendation for the 1% (US+Canada). _{Piotr Konieczny aka Prokonsul Piotrus| reply here} 08:08, 21 July 2026 (UTC)

I manually checked each edit made by my students to catch AI hallucination. But I only asked them to make one edit (and not large-scale editing) so my approach is manageable for now. OhanaUnited^{Talk page} 08:50, 21 July 2026 (UTC)

I gave up on requiring students to do research even before AI era (due to the fact I work with non-native speakers, mamy of whom have abysmally low levels of English fluency). Ironically, MT tools have mostly eliminated that dimension of the problem, but hallucination is a new danger (and when I say mostly, that still means many of my students can produce decent translations, but can't poofread it since they don't understand the output... sigh). _{Piotr Konieczny aka Prokonsul Piotrus| reply here} 08:54, 21 July 2026 (UTC)

It's not a "WikiEdu recommendation". That's just what WikiEdu is themselves doing. Nobody is recommending that individuals do the same thing. WhatamIdoing (talk) 16:56, 21 July 2026 (UTC)