Wikipedia:Administrator elections/December 2025/Candidates/MPGuy2824

Co-nom statement

Allow me to help introduce MPGuy2824, someone who I firmly believe should have been an admin before I became one. They helped me become who I am at NPP, and I owe them a lot for their advice, guidance, and work on related tools at NPP. In addition, they were an NPP coord for several years until they voluntarily stepped down. I should probably also highlight the over 34k patrols and over 84k page curation actions they've taken. They've shown exactly what you want in an admin candidate, knowledge of relevant guidelines, recognition of their own limitations, and a willingness to always be learning and growing. Frankly, I think they've been overly hard on themselves in terms of expectations and should have ran a long time ago.

But if their NPP and tool work wasn't enough in of itself, they've got a very solid content experience resuming to go along with that, in that they've had 14 successful featured list nominations (more than double what I had when I passed!). I've had a number of successful nominations, and I'm proud of each and every one, but I'd have a tough time arguing for any of them over MPGuy2824, who's demonstrated so much in regards to experience that shows they'd make an excellent administrator. I dearly hope you will join me in supporting someone I've been begging to run for a long, long time. Hey man im josh (talk) 21:44, 1 December 2025 (UTC)

Candidate, please indicate acceptance of the nomination here: I accept this nomination. -MPGuy2824 (talk) 08:09, 1 December 2025 (UTC)

Please disclose whether you have ever edited Wikipedia for pay: I have never edited Wikipedia for pay. I have cataloged other accounts that I've used at User:MPGuy2824#Other accounts. -MPGuy2824 (talk)

Optional questions from HurricaneZeta

4. How do you feel about usage of LLMs to generate content (such as new pages) on Wikipedia?

A: Having read a few essays and explanatory pages about this, I basically agree with WP:NEWLLM, but worry about how such text will be reliably identified. Review of LLM-output by the prompter could mean anything from "Yes this looks authoritative", to FA/FL-level scrutiny of prose and sources. LLMs do seem to have some tells and hopefully editors are able to keep track of their drift over time. There are some interesting discussions on this talk page about how to modify the guideline to make it more useful.

5. If you become an administrator, what actions do you think you would take regarding LLM usage?

A: Besides a few areas (e.g. G15 deletions, and blocking/unblocking based on excessive LLM use in discussions), I don't think administrators have any special actions that they can take in relation to LLM usage. I would be comfortable doing those actions when warranted. Admins are also experienced editors though, and in that latter capacity, they can monitor and edit content to keep it sounding less like LLM-output. I haven't really done this yet, but I have noticed that this is a concern among many editors and will be keeping an eye out for this during my watchlist and random patrolling.

Optional question from Conyo14

6. Why did you choose to go through Admin elections instead of RfA?

A: There is a reduction of apprehension when there is a group of people involved, but I’d say the main reason would be to beat my procrastinative tendencies. If I had decided to go the RFA route, the final say on when the transclusion of my nomination page would be mine, and I’m sure there would be thoughts in my head of “maybe it would be better to start this after 2 days (or a week)”. Here, there is a schedule already set and all I had to do was check that there was nothing major going on in my off-wiki life that would be a higher priority. I suspect that we will be getting a majority of our admins via the election route from now on, not necessarily for my reasons.

Optional question from Fade258

7. Which admin action do you feel is the most sensitive and why?

A: "Sensitive" is a somewhat ambiguous term which could mean something to do with privacy/secrecy (e.g. viewing deleted revisions) OR an action that would cause a lot of harm if done badly/maliciously. Going through the list at Special:ListGroupRights#sysop, I think that using over-wide IP ranges during a block could be one sensitive issue. Considering a malicious admin, handing out of advanced permissions to meat puppets would be a cascading problem that would be a huge drain on volunteer time to undo.

Optional question from Jessintime

8. What is your opinion on admin recall generally?

A: Adminship should definitely be revokable, and be revoked when the editor loses the trust of the community. I think that the only way to know if the 25 signature bar for recall is too high or too low is when a few RRFAs have run their course. We haven't had any as yet, although the withdrawn RRFA can be considered as a data point.

Optional question from 11WB

9. Some of your contributions are on tools that reviewers rely on. With the added benefit of having the mop, how would you manage a dilemma where an editor or multiple editors would like a technical change, but that you and other developers believe could be harmful if implemented, rather than beneficial?

A: Anybody with advanced permissions on Wikipedia (like an NPP reviewer or an admin) is expected to implement consensus and not just do things because that’s the way they prefer it. Although there is no explicit user right for a coder/developer, I believe that they should also abide by that same credo. In the case that you lay out, one (or more) developers think a feature change is harmful, while one or more editors think that it is beneficial, the change by definition lacks consensus. An obvious solution that comes to mind would be to bring in a larger group of effected editors to the discussion. If it was something to do with new page patrolling then, by posting on WT:NPPR, and if it involved something that is used by a wider group of editors then WP:VPT might be the right place for that conversation. I do not think that one of the developers also being an admin should have any bearing on the matter.

One example of this consensus-based change is from PageTriage, where a few developers (including myself) thought a particular feature (common tags) was unnecessary and could be removed. Assuming it would be an uncontroversial change we removed it, but faced pushback from reviewers saying that it was a useful and used feature. The feature was added back soon.

There are exceptions to this though, and WP:CONEXCEPT lists some of them.

Optional question from Femke

10. You encounter a slightly promotional article, which you deem isn't unambiguous advertising (not G11). The article has clear signs of AI writing: it has the ** for bolding, a love of m-dashes, flowery conclusions ending each section, the url of one of the three citations is dead on arrival, and the statistic is fabricated. Does this qualify for G15?

A: The dead on arrival citation seems like it clearly meets the condition of "Implausible non-existent references", but it could also be explained by link rot or a typo. More subjective signs as laid out in WP:AISIGNS (like the flowery section conclusions and the bolding via **) can also be used to apply the G15 criterion, but only if the page also meets one of the 3 signs laid out in G15 explicitly. Given that, I feel that I would need to investigate the DOA link further to check for link rot (maybe it is archived somewhere) and plausibility (a scholarly sounding title attributed to a tabloid newspaper, for example). Another investigative thread could include checking the creator's talk page for previous LLM-related warnings/deletions.

Assuming none of the above leads to me gleaning useful data, then as a newer admin this would be in my grey zone, but given that G15 isn't one of the time-sensitive CSD criteria (unlike say copyright infringement or attack pages) and that a bad deletion would be BITEY to a creator, I would fall back on advice given to NPP reviewers: "if you don't know what to do with an article, leave it to another reviewer, add it to your watch list and see what they do with it." In this case, that would mean keeping an eye out for what another admin does to this page, and then discussing with them why they decided the way they did to calibrate my sense of this particular CSD criteria.

Optional question from Vanamonde93

11. Thanks for putting your name forward. Consider the following scenario, which is hypothetical but based on a real example. An editor is topic-banned from CT/SA for persistent additions of unsourced content of a contentious nature. They are subsequently blocked indefinitely for violations of that topic ban. Three years later they file an unblock request, stating "I understand why I was blocked, I will follow the rules, the block is no longer needed". What are your considerations when handling this request, including conditions you may wish to place, and/or questions you may ask?

A: The first thing to check would be how the block was imposed. A block by ArbCom or ANI, for example, needs to be undone via appeals at those locations. For the next few checks, I'm assuming that I wasn't the admin who blocked them in the first place:

• Only saying "I understand why I was blocked" does not show that they really understand: I would point this out to them, expecting an elaboration in reply, since asking a leading question here might be counterproductive.
• What else have they done in the past 3 years? Hopefully it is productive editing in other Wikipedias or in commons, without major warnings received. If they've done only a minor amount of editing, then it is harder to judge a behaviour change, and it might be warranted to ask them to provide an example of the kind of edit they want to make if they are unblocked (2nd chance). Keeping in mind why they were blocked, the edit should be something that is contentious, but has reliable sourcing.

Following this, assuming everything is answered satisfactorily, and this wasn't a block that by ArbCom or ANI, then I'd be willing to unblock them with the topic-ban back in place, which is when I'd ask for input from the administrator who applied the indefinite block. If this was a block imposed by ANI, then I'd be willing to start an appeal for the editor there, with some options including "full unblock, unblock but with the topic ban reimposed (so that they can edit in other areas), OR keep the indefinite block".

Optional question from Docmoates

12. A question above was asked about LLM's. Are you aware of any ways to identify LLM's definitively other than guessing based on the way something is worded? I ask because neurodivergent individuals are often mistaken in their speech for using LLM's^[1][2][3]. I am curious about your thoughts on the matter. Docmoates (talk) 18:31, 8 December 2025 (UTC)

A: