Universal Code of Conduct/Coordinating Committee/Cases/2026/Request for Investigation into Irregularities, Harassment, and Double Standards in the Japanese Wikipedia Bot Approval Process

This case is declined. If you have comments or a request to have it reopened, post a comment on the talk page.

U4C member alert: @U4C: Ajraddatz, Barkeep49, BRPever, Civvì, Dbeef, Ghilt, Ibrahim.ID, Luke081515, Denis Barthel, Ferien, PBradley-WMF. Civvì (talk) 19:44, 25 May 2026 (UTC)

---

碧いうさぎ

Description of the problem - (碧いうさぎ)

Summary of the issue

This request seeks a formal investigation into serious irregularities, harassment, and double standards that occurred during the Japanese Wikipedia Bot Approval Process. The process included 17 days of complete silence, sudden hostile responses, personal attacks, after-the-fact technical criticism, misunderstandings regarding AI usage, and unequal treatment compared to other applicants. All statements and events described below are taken directly from the public discussion on the Bot Request page.

Timeline

17 April 2026 — Submission of Bot Request Bot request page: https://ja.wikipedia.org/wiki/Wikipedia:Bot/使用申請 My Meta user page: https://meta.wikimedia.org/wiki/User:碧いうさぎ

At this time, no administrator or participant requested code disclosure or asked any technical questions.

17 April – 4 May — 17 Days of Complete Silence There were no comments, no questions, and no review activity.

4 May — Test Run Started I began a test run in accordance with the guidelines, as no objections had been raised.

6 May — Sudden Objections and Personal Attacks After 17 days of silence, multiple users and administrators suddenly appeared and posted objections. These comments were not technical; they were subjective criticism and personal remarks.

14 May — Voluntary Disclosure of Source Code I voluntarily published the full source code. I was never asked to provide it. Technical criticism only appeared *after* disclosure, and many concerns were based on misunderstandings.

15 May — Request Rejected and Applicant Blocked A bureaucrat rejected the request, stating “consensus was not achieved,” without addressing the 17-day silence or the hostile behavior of others. No action was taken against users who made derogatory remarks. Only I was blocked.

Statements from involved users

• Nanona15dobato

 - “Promotional”, “AI-generated”, “WP:SILENCE does not exist”, “Your attitude is questionable”, “Oppose”

• Dragoniez（Administrator）

 - “Incoherent AI slop”, “A bot operator who cannot communicate cannot be trusted”,  
   “You are proving yourself impossible to communicate with”, “I don’t feel like replying”, “Oppose”

• W.CC（Administrator）

 - “Feels promotional and AI-generated”, “Page-specific bot design feels promotional”, “I can only oppose”

• 芋羊羹

 - “I have checked the source code”, “It seems it was hard to see”

• Infinite0694

 - “User-Agent spoofing is not acceptable”, “You do not meet the standards”, “I reject this request”

• 柏尾菓子（Bureaucrat, Admin, CheckUser）

 - “I have decided to reject this request”, “Consensus has not been reached”,  
   “You should have explained the technical specifications instead of relying on silence”

Double standards regarding AWB applicant

Another applicant (Balmung0731):

• Began editing without declaring a test run
• Exceeded the 200-edit limit (254 edits)
• Continued mass-editing during review

Talk page: https://ja.wikipedia.org/wiki/利用者‐会話:Balmung0731

No criticism was made toward this user. Administrators reassured them (“No problem”, “Sorry if we caused any concern”).

Meanwhile, my request received personal attacks, hostile language, and after-the-fact technical criticism. Only I was blocked.

Harassment environment

Before any technical review took place, the discussion was dominated by subjective criticism and personal remarks. The review environment effectively became harassment rather than technical evaluation.

Previous attempts at a solution - (碧いうさぎ)

No meaningful attempts were made by administrators to resolve the issue. Concerns were dismissed or ignored. No mediation was offered during the 17 days of silence.

Suggested solutions - (碧いうさぎ)

I respectfully request:

1. A formal investigation into the conduct of the involved administrators and users.
2. Appropriate measures if UCoC violations or misuse of authority are confirmed.
3. A review of structural issues in the Japanese Wikipedia bot approval process.

W.CC

Previous attempts at a solution - W.CC

• I did not make the comments described in the “Statements from involved users” section. The original text of my actual comments is as follows:

 - どのような運転を行うかという申請内容が申請時点で不明確な提案であるため、2週間経過していたとしても仮運用について合意が得られている状況とは思えません。運用内容も拝見しましたが、ウィキペディアの記事に最新版を「１日１回」更新しておく必要性や、Botで行う必要性を理由を説明していただけますでしょうか。--W.CC（会話） 2026年5月6日 (水) 02:32 (UTC)
 - (auto translation) Since the proposal does not clearly specify, at the time of application, what kind of operations will actually be carried out, I do not believe that agreement has been reached regarding provisional operation, even if two weeks have already passed. I have also reviewed the proposed operational details, but could you please explain why it is necessary to update the latest version on the Wikipedia article “once per day,” and why this needs to be done by a bot? - W.CC (talk) 02:32, 6 May 2026 (UTC)

This was the only statement I made. It consisted entirely of reasonable questions regarding the operation of the bot and does not constitute harassment. Furthermore, the applicant’s responses were consistently irrelevant and failed to properly address the issues raised. The bot should not have been approved under JaWp Bot policy.--W.CC (talk) 12:08, 27 May 2026 (UTC)

Suggested solutions - W.CC

The complaint is not based on facts and should be dismissed.--W.CC (talk) 12:08, 27 May 2026 (UTC)

Nanona15dobato

Previous attempts at a solution - Nanona15dobato

On the application page referenced by the complainant, I made three comments (<1>,<2>,<4>) and cast one oppose vote (<3>).

Although the statement presented by 碧いうさぎ regarding my comments is generally coherent, it is also highly fragmented and contains inaccuracies. Therefore, I would like to summarize the comments I made, the reasoning behind them, and provide the supporting quotations below.

• I repeatedly referred to the policy"Wikipedia:Bot"and requested that it be read and understood.^<1><2><3>
• While it is true that technical review is conducted on the application page, such review presupposes that prior consensus regarding the task itself has already been established (1).^<1>
  Accordingly, the discussion had not yet reached the stage of technical review.^<3>
• Regarding approval for trial operation, the policy also states that such determination is to be made by a bureaucrat (2).^<1>
• The guideline cited by the applicant conflicted with this policy, and in principle policies take precedence over guidelines (4).^<3>
  Furthermore, 碧いうさぎ's application did not specify the bot approval type, which itself could be considered a defect in the application.^<1>
• Given that the source code had only just been presented, the statement that "the code had not been reviewed for 17 days" was not plausible.^<3><4>

Next, regarding the AI-generated comments: 碧いうさぎ explicitly acknowledged using AI in their responses (5), but I do not consider the use of AI itself to be problematic.

• The issue was not the use of AI, but rather that comments appeared to have been posted without verification.^<2><3>
• As evidence of this, 碧いうさぎ referred to a Japanese Wikipedia page titled "WP:SILENCE"; however, no such page exists on jawp, and the reference appears to have instead pointed to another project such as the English Wikipedia.^<2>
• I also considered it problematic that meaningful dialogue was not being established due to the manner in which AI was being used.^<2><3>

---
Supporting quotations

w:ja:Wikipedia:Bot（policy）

(1) "Requests for bot approval are limited to discussion of the bot’s functionality. Discussions regarding whether the task itself is necessary or appropriate should be conducted separately on the appropriate page in advance."/「Botの使用申請では、Botの機能に関する審議のみを行います。作業そのものが必要であるか、妥当であるかに関する議論は、別途適切なページで事前に行ってください。」

(2) "A bureaucrat determines whether prior consensus for trial operation has been obtained. After consensus is reached, trial operation should be conducted to demonstrate that there are no issues with speed or editing behavior."/「仮運用の事前合意が得られたか、ビューロクラットが判断します。合意後に仮運用を行い、速度と編集内容に問題がないことを実証してください。」

(3) "For Type C bots as well, prior consensus for trial operation is not required when processing tasks that have already been submitted to WP:BOTREQ."/「種別CについてもWP:BOTREQが出されている案件を処理する場合は仮運用の事前合意は不要です。」

w:ja:Wikipedia:方針とガイドライン#guides（policy）

(4) "Where guidelines conflict with policies, policies normally take precedence."/「ガイドラインが方針と衝突する場合には、通常は方針を優先します。」

:w:ja:Wikipedia:Bot/使用申請#Wazejapan_Country_Manager_の使用申請（BRFA）

(5) "The reason I had been responding with AI was to confirm whether you had forgotten Wikipedia’s rules; I was using AI’s knowledge to check this with you."/「今までAIで返答していたのは、あなた方がウィキペディアのルールを忘れていないか確認するために、AIの知識を使ってあなた方に確認を送っていたものです。」

---

As such, my comments and subsequent discussions were based on policy rather than subjective impressions. In particular, I referred to Wikipedia:Bot multiple times even before expressing opposition.

In addition, the statement describing my comments as "Promotional" is inaccurate, as I made no references whatsoever to promotion in that discussion.

Furthermore, Balmung0731-AWB was a Type C bot application. In such cases, prior consensus is already established through BOTREQ, and therefore trial operation may proceed without a bureaucrat’s determination (3). Accordingly, there was no double standard between that application and the present case regarding prior consensus.--nanona(15) 04:47, 28 May 2026 (UTC)

Suggested solutions - Nanona15dobato

• My comments, as well as those made by other participants and the resulting determinations on the application page, were legitimate and valid.
• This complaint is not based on the facts and should therefore be dismissed.--nanona(15) 04:47, 28 May 2026 (UTC)

Other feedback

For people who are not parties, the following rules apply:

• Comments/replies may not be longer the 500 words and may not include more than 25 diffs/links. The U4C may, if asked, grant additional words or diffs/links.
• Comments/replies are permitted only in your own section
• Contributions that do not help clarify the matter can be removed
• All accusations and claims must be supported with diffs/links

Other feedback (USSR-Slav)

I would like to provide some context as an uninvolved observer.

There were substantial procedural and technical reasons for concern. The bot request described the task only as updating the latest version information of the software “Waze”, and it was not clear from the initial request which bot task category under ja:Wikipedia:Bot#作業種別 it corresponded to. W.CC therefore asked why the proposal was necessary as a bot task and stated that the request was unclear at the time of filing(l.33 to 34).

There were also concrete technical objections. Dragoniez and Infinite0694 raised concerns such as the mismatch between the declared “fully automatic” operation and the visible source code, Google Play scraping, User-Agent spoofing, fragile parsing logic, insufficient error handling, and the possibility that the code had not been adequately reviewed before deployment. These are not merely personal attacks; they are relevant to whether a bot operator and the proposed task met the local requirements(2026-5-15-13:41,2026-5-15-15:25).

As additional context, the version of ja:Waze edited by 碧いうさぎ, ja:特別:差分/109522717, may also be relevant. That version appears to contain unusual style features, including excessive emoji use and unnatural wording, and there seem to be broader sourcing and original-research concerns. This may explain why some participants suspected LLM use or inadequate review.

The block was made by Mt.Asahidake for the reason of "a period for reviewing various policies: WP:BOT, WP:IDIDNTHEARTHAT". I do not have enough information to conclude whether the block was appropriate, but this context may be relevant when assessing whether the block was policy-based or punitive.

The comparison with Balmung0731-AWB should also be treated carefully. Balmung0731-AWB was described as a re-application for a bot that had previously held a bot flag around 2010, and its proposed work involved conventional AWB-assisted tasks such as link, category, template, and typo/style corrections. This makes it materially different from the Wazejapan Country Manager request, which proposed a new page-specific task involving automatic version-data updates from Google Play.--_USSR-Slav (СССР-Слав) (talk) 10:45, 26 May 2026 (UTC)

To the members of the Japanese Wikipedia community,

First of all, this is not a place to declare that “Usagi is at fault.” The purpose of this discussion is to examine whether the issues that have actually occurred on the Japanese Wikipedia are appropriate under its current operational practices. Please read the topic carefully and discuss it based on its content.

At present, the article on Waze still contains the following incorrect statement:

Symbian, Maemo, Windows Mobile, and BlackBerry OS (beta only) are supported.

These operating systems no longer exist, and Waze does not support them. If you are going to revert an edit, please verify the facts properly before doing so.

Reverting simply because “you dislike Usagi” is not acceptable. Please check the current situation and base your actions on accurate information.

If this situation continues, it will inevitably lead to questions such as: “Is spreading incorrect information the rule of the Japanese Wikipedia?” This would be a serious problem for the entire project.

What I would like to confirm here is whether providing and maintaining accurate information is not a fundamental responsibility of those involved in Wikipedia. I ask that you refrain from shifting the discussion and instead address the issue in accordance with the stated topic. 碧いうさぎ (talk) 02:17, 2 June 2026 (UTC)

Discussion between the involved parties and the U4C members

Only the involved parties and U4C members may edit in this section.

• @碧いうさぎ: please notify all involved parties on jawiki, thank you. --Civvì (talk) 19:52, 25 May 2026 (UTC)
• @碧いうさぎ: Which of the critical remarks do you see as sanctionable? --Ghilt (talk) 21:19, 25 May 2026 (UTC)

  Thank you for your question.

  First, I am not in a position to determine which specific statements should be subject to sanctions. My role is to present the facts — the statements, actions, and timeline that actually occurred — and leave the evaluation to the U4C.

  UCoC 2.1: Insults and Personal Attacks

  • Expressions such as “AI slop” and “impossible to communicate with”
  • Remarks that demean the applicant’s character or competence

  UCoC 2.3: Unfounded or Misguided Criticism

  • Criticism based on misunderstandings about AI usage
  • Accusations based on a misreading of WP:SILENCE

  UCoC 2.2: Abuse or Misuse of Authority

  • A 17‑day period of silence followed by a sudden rejection without explanation
  • Failure to address insulting remarks by others while blocking only the applicant

  UCoC 2.6: Unequal or Selective Enforcement

  • Lenient treatment toward other bot applicants, contrasted with disproportionately harsh treatment toward me

  These points are not assertions of guilt, nor am I demanding sanctions. They are a factual organization of events intended to assist the U4C in its neutral evaluation. The final determination of whether any statement constitutes a UCoC violation rests entirely with the U4C.

  Concerns Regarding Compliance with Wikipedia Policies In addition, the series of statements and actions documented in my submission may also serve as evidence that some of the involved users themselves failed to follow Wikipedia’s fundamental policies and the UCoC.

  Examples include:

  • Use of insulting or demeaning expressions
  • Criticism based on misunderstandings or incorrect assumptions
  • A sudden rejection after prolonged silence
  • Selective enforcement in which only the applicant was sanctioned

  These behaviors appear inconsistent with core Wikipedia principles such as treating others with respect, assuming good faith, and ensuring fair and impartial administrative actions.

  In this sense, the statements and actions at issue may indicate that the individuals involved were themselves disregarding established rules and standards.

  While I leave all final judgments to the U4C, I believe these points are important for understanding the context of this case, and I have organized them accordingly. 碧いうさぎ (talk) 21:25, 29 May 2026 (UTC)

  i don't see personal attacks there, just criticism. That is not automatically a personal attack. And you were also informed about the reason for the long duration and the rejection: missing prerequisites. And on the bot requests of other people: there is no equality in wrongdoing. That means, you (or anyone else) do not get a right to skip requirements if others skipped, but you can ask that other bot requests are treated according to the rules. Nothing more. Ghilt (talk) 22:23, 29 May 2026 (UTC)

• I think (A review of structural issues in the Japanese Wikipedia bot approval process) is out of U4C scope, We don't exercise any oversight over technical or administrative decisions; what concerns us is only: abuse of power (violate the local policy) or behavioral violations.--Ibrahim.ID (talk) 05:46, 26 May 2026 (UTC)

• @碧いうさぎ: it might be a translation issue but it does not seem to me that your bot request was submitted following the indications written here ja:Wikipedia:Bot/使用申請. I guess this has something to do with how the approval process did (not) proceed and some of the remarks on that page? --Civvì (talk) 08:59, 26 May 2026 (UTC)

  Regarding your question about the approval process: I did not begin the test run arbitrarily. Another applicant had already started their test run in a similar situation, and since there had been no comments from administrators for more than two weeks, I assumed that administrators might simply be unavailable at that time.

  Based on that context, and following advice I received from AI tools suggesting that it would be better to clearly announce the start of the test run, I proceeded accordingly. My intention was to follow the written procedure as closely as possible, not to bypass it.

  For this reason, it was difficult for me to understand why procedural concerns were raised only after such a long period of silence. This difference between the documented process and how the situation was actually handled contributed to my confusion.

  If further clarification is needed, I will be glad to provide more details. 碧いうさぎ (talk) 22:26, 29 May 2026 (UTC)

@碧いうさぎ: I am talking about the steps described in this page ja:Wikipedia:Bot/使用申請, the second point is "Identify your bot on its user page using the {{bot}} template", the user page of your bot is still empty ja:利用者:Wazejapan_Country_Manager, that might be the main reason why the entire approval process has neither properly started nor made any meaningful progress. Carefully reading the instructions is never a waste of time; it can prevent a lot of frustration and generally saves time for everyone involved in a process. --Civvì (talk) 12:47, 31 May 2026 (UTC)

Thank you for your comment. Regarding the point that the bot’s user page is empty, the page I had previously created (ja:利用者:Wazejapan_Country_Manager) may have been deleted by an administrator when the request was rejected. If that was the case, then adding the {{bot}} template or completing the page might not have been possible at that stage.

Based on the wording of Wikipedia:Bot/使用申請, my understanding was that the {{bot}} template is generally added after approval. Because of that, I am not entirely sure how much this point may have influenced the progress of the request.

In addition, the bot approval process seems to depend heavily on administrator action. If there had been any issues in my procedure, it seems possible that receiving feedback within the first one or two weeks would have been the usual flow. In my case, there was no response for more than two weeks, and only afterward was I told that the process “had not started”, so I may not have had any opportunity to correct anything in advance.

Given this structure, I also wonder whether, in cases where something is missing or insufficient, it would normally be the administrator who points out what needs to be corrected. That seems to be the usual pattern in many other processes on Wikipedia.

For these reasons, it seems possible that the situation may have been influenced more by procedural circumstances than by anything on my side. 碧いうさぎ (talk) 14:07, 31 May 2026 (UTC)

@碧いうさぎ: the page was never created, there are no deletions in the log and no creations of that page in your contributions. And the steps in that page are explained in a really clear way, requesting approval is point 7 meaning that what is described in the previous points needs to be done "before" that. I would warmly recommend gaining some more hands-on editing experience on jawiki before moving on with bot-related work. --Civvì (talk) 16:22, 31 May 2026 (UTC)

The original issue here was whether the actions and statements of the Japanese Wikipedia administrators were appropriate. However, the discussion has shifted into claiming that “I was at fault,” which is not the point and represents a clear deviation from the actual topic.

I would like the evaluation to be based on the statements and actions recorded on the case page. From what is written there, it should be evident that some administrators acted toward me with hostility or bad faith. I ask that the review be conducted based on the documented facts, without being influenced by the administrators’ own explanations. 碧いうさぎ (talk) 02:51, 2 June 2026 (UTC)

You do not control the topic. We have shown you that you did not meet the requirements and ignored the feedback. Ghilt (talk) 09:16, 2 June 2026 (UTC)

For example, are we to understand that if a beginner makes a minor mistake in a bot request, it is considered “normal” under global Wikipedia standards to leave them waiting for 17 days and then subject them to hostile remarks?

Furthermore, even after the response period defined by Wikipedia has passed, are users still expected to wait—one year, ten years, however long it takes—until an administrator eventually appears? If this is truly the standard practice, it raises serious concerns.

And if this really is the accepted norm within Wikipedia, then shouldn’t the Japanese Wikipedia clearly state on its main page, in Japanese, something like: “Please wait one or two years until an administrator responds”? Users deserve to know such rules upfront to avoid misunderstandings.

If this is indeed considered proper practice, I respectfully request that the Japanese Wikipedia administrators be instructed to clearly communicate this standard to their users. 碧いうさぎ (talk) 03:06, 2 June 2026 (UTC)

That didn't qualify as "a minor mistake" but as a series of mistakes. You also didn't improve your request. And you fail to acknowledge that, which is problematic. Ghilt (talk) 09:19, 2 June 2026 (UTC)

The statement you provided lacks a clear subject, so it is not apparent which requirements or which feedback you are referring to. As a result, I am unable to understand what your comment is addressing. If you could specify these points, I would be able to respond accurately. 碧いうさぎ (talk) 09:29, 2 June 2026 (UTC)

My apologies. I later realized that the text I was translating was not the correct one. With that clarified, I would like to ask: Under normal circumstances, when a newcomer makes a mistake, isn’t it standard practice for administrators to point out the issue and guide them appropriately? In this case, however, I was left waiting for a long period, received no guidance, and was even subjected to hostile remarks. Is such treatment considered “normal” within Wikipedia?

And is it also regarded as normal practice for members to shift the point of discussion during the debate? 碧いうさぎ (talk) 09:57, 2 June 2026 (UTC)

U4C decision

Only U4C members may edit in this section.

Decline votes

• The criticism of the bot request seems acceptable and valid to me. Therefore, i don't see a sanctionable UCoC violation here. Please try to work on the issues raised with your bot. Ghilt (talk) 22:31, 29 May 2026 (UTC)
• I do not see UCoC violations. I would also like to invite the user to pay closer attention to pages describing procedures and required steps. --Civvì (talk) 12:57, 31 May 2026 (UTC)
• --BRP ^ever 05:49, 1 June 2026 (UTC)
• I agree with my colleagues. Barkeep49 (talk) 18:05, 1 June 2026 (UTC)
• I don't see clear UCoC violations, Agree with my colleagues --Ibrahim.ID (talk) 13:26, 3 June 2026 (UTC)

Motions

U4C members may propose motions to resolve the case or as a temporary measure during the case.

Updates

This section is used only by U4C members and official designees (including WMF staff who support the U4C) to provide updates about the request.

We have seen this, on behalf of the U4C --Civvì (talk) 19:52, 25 May 2026 (UTC)

With 5 votes the request is declined. On behalf of the U4C, --Civvì (talk) 10:16, 4 June 2026 (UTC)