[{"id":"591d3e6d-448e-41c9-b9c8-b381f79c308d","thread_id":"674061bf-35fc-496d-8b54-17b25db50fd7","created_at":"2026-02-19T23:02:26Z","branch_list":["00000000-0000-4000-0000-000000000000","076941ae-ab29-4129-8508-fc63755bb3e7"],"state":"done","prompt":"Summarise the arguments and outcome of this German Wikipedia policy discussion: https://de.wikipedia.org/wiki/Wikipedia:Meinungsbilder/KI-Verbot. Are there any useful takeaways for similar issues on English Wikipedia?","references_html":"<ol data-ref-list>\n \n <li>\n <a href=\"https://en.wikipedia.org/wiki/Wikipedia:Artificial_intelligence\" >Wikipedia:Artificial intelligence - Wikipedia</a>\n <span class=\"__domain-name\">en.wikipedia.org</span>\n <div class=\"trigger\" tabindex=\"0\">\n <span class=\"contribution\" style=\"background: linear-gradient(to right, var(--primary-70) 27%, transparent 27%)\">27%</span>\n <div class=\"tooltip top end tooltip_base tw-auto\">\n This percentage indicates how much a source has contributed to the answer.\n </div>\n </div>\n </li>\n \n <li>\n <a href=\"https://de.wikipedia.org/wiki/Wikipedia:Meinungsbilder/KI-Verbot\" >Wikipedia:Meinungsbilder/KI-Verbot – Wikipedia</a>\n <span class=\"__domain-name\">de.wikipedia.org</span>\n <div class=\"trigger\" tabindex=\"0\">\n <span class=\"contribution\" style=\"background: linear-gradient(to right, var(--primary-70) 25%, transparent 25%)\">25%</span>\n <div class=\"tooltip top end tooltip_base tw-auto\">\n This percentage indicates how much a source has contributed to the answer.\n </div>\n </div>\n </li>\n \n <li>\n <a href=\"https://en.wikipedia.org/wiki/Wikipedia:AI-generated_content\" >Wikipedia:AI-generated content</a>\n <span class=\"__domain-name\">en.wikipedia.org</span>\n <div class=\"trigger\" tabindex=\"0\">\n <span class=\"contribution\" style=\"background: linear-gradient(to right, var(--primary-70) 19%, transparent 19%)\">19%</span>\n <div class=\"tooltip top end tooltip_base tw-auto\">\n This percentage indicates how much a source has contributed to the answer.\n </div>\n </div>\n </li>\n \n <li>\n <a href=\"https://en.wikipedia.org/wiki/Artificial_intelligence_in_Wikimedia_projects\" >Artificial intelligence in Wikimedia projects - Wikipedia</a>\n <span class=\"__domain-name\">en.wikipedia.org</span>\n <div class=\"trigger\" tabindex=\"0\">\n <span class=\"contribution\" style=\"background: linear-gradient(to right, var(--primary-70) 12%, transparent 12%)\">12%</span>\n <div class=\"tooltip top end tooltip_base tw-auto\">\n This percentage indicates how much a source has contributed to the answer.\n </div>\n </div>\n </li>\n \n <li>\n <a href=\"https://en.wikipedia.org/wiki/Wikipedia:WikiProject_AI_Cleanup/Policies\" >Wikipedia:WikiProject AI Cleanup/Policies - Wikipedia</a>\n <span class=\"__domain-name\">en.wikipedia.org</span>\n <div class=\"trigger\" tabindex=\"0\">\n <span class=\"contribution\" style=\"background: linear-gradient(to right, var(--primary-70) 7%, transparent 7%)\">7%</span>\n <div class=\"tooltip top end tooltip_base tw-auto\">\n This percentage indicates how much a source has contributed to the answer.\n </div>\n </div>\n </li>\n \n <li>\n <a href=\"https://en.wikipedia.org/wiki/Wikipedia:Signs_of_AI_writing\" >Wikipedia:Signs of AI writing</a>\n <span class=\"__domain-name\">en.wikipedia.org</span>\n <div class=\"trigger\" tabindex=\"0\">\n <span class=\"contribution\" style=\"background: linear-gradient(to right, var(--primary-70) 4%, transparent 4%)\">4%</span>\n <div class=\"tooltip top end tooltip_base tw-auto\">\n This percentage indicates how much a source has contributed to the answer.\n </div>\n </div>\n </li>\n \n <li>\n <a href=\"https://en.wikipedia.org/wiki/Wikipedia:AI_image_use\" >Wikipedia:AI image use</a>\n <span class=\"__domain-name\">en.wikipedia.org</span>\n <div class=\"trigger\" tabindex=\"0\">\n <span class=\"contribution\" style=\"background: linear-gradient(to right, var(--primary-70) 3%, transparent 3%)\">3%</span>\n <div class=\"tooltip top end tooltip_base tw-auto\">\n This percentage indicates how much a source has contributed to the answer.\n </div>\n </div>\n </li>\n \n <li>\n <a href=\"https://en.wikipedia.org/wiki/Wikipedia:Image_use_policy\" >Wikipedia:Image use policy</a>\n <span class=\"__domain-name\">en.wikipedia.org</span>\n <div class=\"trigger\" tabindex=\"0\">\n <span class=\"contribution\" style=\"background: linear-gradient(to right, var(--primary-70) 3%, transparent 3%)\">3%</span>\n <div class=\"tooltip top end tooltip_base tw-auto\">\n This percentage indicates how much a source has contributed to the answer.\n </div>\n </div>\n </li>\n \n</ol>\n","references_md":"[^1]: [Wikipedia:Artificial intelligence - Wikipedia](https://en.wikipedia.org/wiki/Wikipedia:Artificial_intelligence) (27%)\n[^2]: [Wikipedia:Meinungsbilder/KI-Verbot – Wikipedia](https://de.wikipedia.org/wiki/Wikipedia:Meinungsbilder/KI-Verbot) (25%)\n[^3]: [Wikipedia:AI-generated content](https://en.wikipedia.org/wiki/Wikipedia:AI-generated_content) (19%)\n[^4]: [Artificial intelligence in Wikimedia projects - Wikipedia](https://en.wikipedia.org/wiki/Artificial_intelligence_in_Wikimedia_projects) (12%)\n[^5]: [Wikipedia:WikiProject AI Cleanup/Policies - Wikipedia](https://en.wikipedia.org/wiki/Wikipedia:WikiProject_AI_Cleanup/Policies) (7%)\n[^6]: [Wikipedia:Signs of AI writing](https://en.wikipedia.org/wiki/Wikipedia:Signs_of_AI_writing) (4%)\n[^7]: [Wikipedia:AI image use](https://en.wikipedia.org/wiki/Wikipedia:AI_image_use) (3%)\n[^8]: [Wikipedia:Image use policy](https://en.wikipedia.org/wiki/Wikipedia:Image_use_policy) (3%)\n","reply":"<details><summary>Gathered key details from:<p></p>\n<ul>\n<li><a href=\"https://de.wikipedia.org/wiki/Wikipedia:Meinungsbilder/KI-Verbot\" target=\"_blank\">Wikipedia:Meinungsbilder/KI-Verbot – Wikipedia</a></li></ul></summary>\n\n<h3>Query</h3>\n<p>Summarize the arguments for and against the 'KI-Verbot' (AI ban) and the outcome of the discussion.</p>\n<h3>Response</h3>\n<p>The 'KI-Verbot' (AI ban) discussion aimed to determine whether the use of Artificial Intelligence, particularly Large Language Models (LLMs), should be largely prohibited in Wikipedia.</p>\n<p>Arguments for the ban emphasized preserving Wikipedia as a project of human-written content, preventing the influx of AI-generated \"slop\" (AI-Müll), and maintaining Wikipedia's reputation as a beacon of human-authored content. The proponents argued that now is the opportune time to establish this, as not too much AI content has infiltrated yet.</p>\n<p>Arguments against the ban highlighted the lack of clear recognition criteria for AI-generated content, the absence of detailed procedures for handling such content and repeat offenders, and the risk of false positives leading to unfair accusations. Opponents also raised concerns that a ban might drive AI use underground, exclude useful future AI tools, focus on the method rather than quality, and make Wikipedia seem outdated. They pointed to other projects integrating AI and suggested focusing on quality standards rather than banning tools. Many also argued that such a significant rule change required a qualified majority (2/3) rather than a simple majority.</p>\n<p>The outcome of the discussion was that the Meinungsbild (opinion poll/discussion) was formally accepted, and the proposal for the new guideline page \"Wikipedia:Künstliche Intelligenz\" was also accepted with the required majority. This means a new guideline page will be created and linked, establishing rules regarding the use of AI in Wikipedia. A total of 350 generally eligible users cast their votes.</p>\n</details>\n<details><summary>Searched with Kagi<p></p>\n<ul>\n<li>English Wikipedia policy on AI generated content</li>\n<li>English Wikipedia discussions on AI content</li></ul></summary><br>\nSources (20):<br>\n<strong>Wikipedia:AI-generated content</strong><br>\n<a href=\"https://en.wikipedia.org/wiki/Wikipedia:AI-generated_content\" target=\"_blank\">https://en.wikipedia.org/wiki/Wikipedia:AI-generated_content</a><br>\nVarious types of computer algorithms, neural networks, machine learning, and approaches sometimes termed artificial intelligence are used to assist humans in generating creative material. Each editor of Wikipedia is responsible for ensuring that everything they submit to Wikipedia meets its policies, guidelines, standards and best practices, and that any and all facts and claims are accurate.\n\n<p><strong>Wikipedia:Artificial intelligence - Wikipedia</strong><br>\n<a href=\"https://en.wikipedia.org/wiki/Wikipedia:Artificial_intelligence\" target=\"_blank\">https://en.wikipedia.org/wiki/Wikipedia:Artificial_intelligence</a><br>\n Artificial intelligence (AI) is used on a number of Wikipedia and Wikimedia projects. This may be directly involved with creation of text content, or in support roles related to evaluating article quality, adding metadata, or generating images. As with any machine-generated content, care must be used when employing AI at scale or in applying it where the community consensus is to exercise more ...</p>\n<p><strong>Artificial intelligence in Wikimedia projects - Wikipedia</strong><br>\n<a href=\"https://en.wikipedia.org/wiki/Artificial_intelligence_in_Wikimedia_projects\" target=\"_blank\">https://en.wikipedia.org/wiki/Artificial_intelligence_in_Wikimedia_projects</a><br>\n [1] Some applications of artificial intelligence, like using large language models to create new articles from scratch, have been more controversial than others for the Wikipedia community. In August 2025, Wikipedia adopted a policy that allowed editors to nominate suspected AI-generated articles for speedy deletion.</p>\n<p><strong>Wikipedia:WikiProject AI Cleanup/Policies - Wikipedia</strong><br>\n<a href=\"https://en.wikipedia.org/wiki/Wikipedia:WikiProject_AI_Cleanup/Policies\" target=\"_blank\">https://en.wikipedia.org/wiki/Wikipedia:WikiProject_AI_Cleanup/Policies</a><br>\n Wikipedia:Talk page guidelines § LLM-generated (WP:AITALK), a guideline entry allowing the striking or collapsing of comments that are obviously generated by an LLM or similar AI technology The following are not policies or guidelines, but still have some significance in this context:</p>\n<p><strong>Wikipedia:Signs of AI writing</strong><br>\n<a href=\"https://en.wikipedia.org/wiki/Wikipedia:Signs_of_AI_writing\" target=\"_blank\">https://en.wikipedia.org/wiki/Wikipedia:Signs_of_AI_writing</a><br>\n Leads treating Wikipedia lists or broad article titles as proper nouns. In AI-generated articles about topics with a title that is not a proper name, such as a ...</p>\n<p><strong>Generative artificial intelligence - Wikipedia</strong><br>\n<a href=\"https://en.wikipedia.org/wiki/Generative_artificial_intelligence\" target=\"_blank\">https://en.wikipedia.org/wiki/Generative_artificial_intelligence</a><br>\n Generative AI models have been trained on copyrighted works without the rightholders' permission. ... Many generative AI systems use large-scale data centers ...</p>\n<p><strong>Wikipedia:Wikipedia Signpost/2024-10-19/Recent research</strong><br>\n<a href=\"https://en.wikipedia.org/wiki/Wikipedia:Wikipedia_Signpost/2024-10-19/Recent_research\" target=\"_blank\">https://en.wikipedia.org/wiki/Wikipedia:Wikipedia_Signpost/2024-10-19/Recent_research</a><br>\nOct 19, 2024 \"As many as 5%\" of new English Wikipedia articles \"contain significant AI-generated content\" Figure 1 from the paper: \"Using two tools, GPTZero and Binoculars, ...</p>\n<p><strong>Wikipedia:Image use policy</strong><br>\n<a href=\"https://en.wikipedia.org/wiki/Wikipedia:Image_use_policy\" target=\"_blank\">https://en.wikipedia.org/wiki/Wikipedia:Image_use_policy</a><br>\n AI-generated images Most images wholly generated by AI should not be used in mainspace, especially for named people and in technical or scientific subjects ...</p>\n<p><strong>Wikipedia:No original research</strong><br>\n<a href=\"https://en.wikipedia.org/wiki/Wikipedia:No_original_research\" target=\"_blank\">https://en.wikipedia.org/wiki/Wikipedia:No_original_research</a><br>\n This page documents an English Wikipedia policy. ... Most images wholly generated by AI should not be used in mainspace; see WP:Image use policy § AI-generated ...</p>\n<p><strong>Wikipedia:AI image use</strong><br>\n<a href=\"https://en.wikipedia.org/wiki/Wikipedia:AI_image_use\" target=\"_blank\">https://en.wikipedia.org/wiki/Wikipedia:AI_image_use</a><br>\n Images generated wholly by AI should generally not be used to illustrate Wikipedia articles. Exceptions to this rule exist when the image is notable in itself.</p>\n<p><strong>Wikipedia:Signs of AI writing</strong><br>\n<a href=\"https://en.wikipedia.org/wiki/Wikipedia:Signs_of_AI_writing\" target=\"_blank\">https://en.wikipedia.org/wiki/Wikipedia:Signs_of_AI_writing</a><br>\n This is a list of writing and formatting conventions typical of AI chatbots such as ChatGPT, with real examples taken from Wikipedia articles and drafts. It is a field guide to help detect undisclosed AI-generated content on Wikipedia: while some of the signs may be broadly applicable, some may not apply in a non-Wikipedia context.</p>\n<p><strong>Wikipedia talk:AI-generated content - Wikipedia</strong><br>\n<a href=\"https://en.wikipedia.org/wiki/Wikipedia_talk:AI-generated_content\" target=\"_blank\">https://en.wikipedia.org/wiki/Wikipedia_talk:AI-generated_content</a><br>\n The following is a closed discussion of a requested move. Please do not modify it. Subsequent comments should be made in a new section on the talk page. Editors desiring to contest the closing decision should consider a move review after discussing it on the closer's talk page. No further edits should be made to this discussion. The result of the move request was: page moved. Andrewa (talk) 01 ...</p>\n<p><strong>Wikipedia:AI-generated content</strong><br>\n<a href=\"https://en.wikipedia.org/wiki/Wikipedia:AI-generated_content\" target=\"_blank\">https://en.wikipedia.org/wiki/Wikipedia:AI-generated_content</a><br>\n Wikipedia:AI-generated content ... This page in a nutshell: Computer algorithms are increasingly used to augment human-generated material and Wikipedia editors are fully responsible for ensuring the quality of all content submitted by such means and the veracity any claims made about it.</p>\n<p><strong>Artificial intelligence - Wikipedia</strong><br>\n<a href=\"https://en.wikipedia.org/wiki/Artificial_intelligence\" target=\"_blank\">https://en.wikipedia.org/wiki/Artificial_intelligence</a><br>\n Artificial intelligence (AI) is the capability of computational systems to perform tasks typically associated with human intelligence, such as learning, reasoning, problem-solving, perception, and decision-making. It is a field of research in computer science that develops and studies methods and software that enable machines to perceive their environment and use learning and intelligence to ...</p>\n<p><strong>Artificial intelligence in Wikimedia projects - Wikipedia</strong><br>\n<a href=\"https://en.wikipedia.org/wiki/Artificial_intelligence_in_Wikimedia_projects\" target=\"_blank\">https://en.wikipedia.org/wiki/Artificial_intelligence_in_Wikimedia_projects</a><br>\n Artificial intelligence and machine learning have long been used in Wikimedia projects, primarily to help edit existing articles. [1] Some applications of artificial intelligence, like using large language models to create new articles from scratch, have been more controversial than others for the Wikipedia community. In August 2025, Wikipedia adopted a policy that allowed editors to nominate ...</p>\n<p><strong>Wikipedia:Artificial intelligence</strong><br>\n<a href=\"https://en.wikipedia.org/wiki/Wikipedia:Artificial_intelligence\" target=\"_blank\">https://en.wikipedia.org/wiki/Wikipedia:Artificial_intelligence</a><br>\n Artificial intelligence (AI) is used on a number of Wikipedia and Wikimedia projects. This may be directly involved with creation of text content, or in support roles related to evaluating article quality, adding metadata, or generating images. As with any machine-generated content, care must be used when employing AI at scale or in applying it where the community consensus is to exercise more ...</p>\n<p><strong>Artificial intelligence content detection - Wikipedia</strong><br>\n<a href=\"https://en.wikipedia.org/wiki/Artificial_intelligence_content_detection\" target=\"_blank\">https://en.wikipedia.org/wiki/Artificial_intelligence_content_detection</a><br>\n Artificial intelligence detection software aims to determine whether some content (text, image, video or audio) was generated using artificial intelligence (AI ...</p>\n<p><strong>Generative artificial intelligence - Wikipedia</strong><br>\n<a href=\"https://en.wikipedia.org/wiki/Generative_artificial_intelligence\" target=\"_blank\">https://en.wikipedia.org/wiki/Generative_artificial_intelligence</a><br>\n Generative artificial intelligence, also known as generative AI or GenAI, is a subfield of artificial intelligence that uses generative models to generate ...</p>\n<p><strong>Talk:AI-generated content on Wikipedia</strong><br>\n<a href=\"https://en.wikipedia.org/wiki/Draft_talk:AI_generated_content_on_Wikipedia\" target=\"_blank\">https://en.wikipedia.org/wiki/Draft_talk:AI_generated_content_on_Wikipedia</a><br>\n You can use this page to start a discussion with others about how to improve the \"AI-generated content on Wikipedia\" page. Start a discussion · Add topic · Last ...</p>\n<p><strong>Wikipedia:Wikipedia Signpost/2024-10-19/Recent research - Wikipedia</strong><br>\n<a href=\"https://en.wikipedia.org/wiki/Wikipedia:Wikipedia_Signpost/2024-10-19/Recent_research\" target=\"_blank\">https://en.wikipedia.org/wiki/Wikipedia:Wikipedia_Signpost/2024-10-19/Recent_research</a><br>\nOct 19, 2024 A new paper titled \"The Rise of AI-Generated Content in Wikipedia\" [1] estimates \"that 4.36% of 2,909 English Wikipedia articles created in August 2024 contain significant AI-generated content\" In more detail, the authors used two existing AI detectors (GPTZero and Binoculars), which \"reveal a marked increase in AI-generated content in recent [ly created] pages compared to those from before ...</p>\n</details><details><summary>Thinking</summary>\n<p>The user wants a summary of the German Wikipedia \"KI-Verbot\" (AI ban) Meinungsbild, plus takeaways for English Wikipedia. Let me synthesize the information I have.</p>\n</details>\n<h2>German Wikipedia: KI-Verbot Meinungsbild</h2>\n<h3>Overview</h3>\n<p>The <em>Meinungsbild</em> (a formal community opinion poll, roughly analogous to an English Wikipedia RfC but with binding force) proposed to largely prohibit the use of AI — particularly large language models — in German Wikipedia, and to create a new guideline page, <strong>Wikipedia:Künstliche Intelligenz</strong>, codifying rules on AI use <sup class=\"trigger\"><a href=\"https://de.wikipedia.org/wiki/Wikipedia:Meinungsbilder/KI-Verbot#:~:text=The%20%27KI-Verbot%27%20%28AI,cast%20their%20votes.\" target=\"_blank\">2</a><span class=\"tooltip top center citation\"><a href=\"https://de.wikipedia.org/wiki/Wikipedia:Meinungsbilder/KI-Verbot#:~:text=The%20%27KI-Verbot%27%20%28AI,cast%20their%20votes.\"><span class=\"source\"><img alt=\"\" src=\"https://p.kagi.com/proxy/favicons?c=lQ30-cxV_E2Vdyqb6hrbuYgId7qu7hecjO7o1T7aqQDFc7XpcLttL1Ryala-GNHJAQ-fjt8WMTZhLN6U1wzeZKYM0S0jk4QmuuY129Xs3iY%3D\"> de.wikipedia.org</span><b>2. Wikipedia:Meinungsbilder/KI-Verbot – Wikipedia</b></a><span class=\"passage\">The 'KI-Verbot' (AI ban) discussion aimed to determine whether the use of Artificial Intelligence, particularly Large Language Models (LLMs), should be largely prohibited in Wikipedia. Arguments for the ban emphasized preserving Wikipedia as a project of human-written content, preventing the influx of AI-generated \"slop\" (AI-Müll), and maintaining Wikipedia's reputation as a beacon of human-authored content. The proponents argued that now is the opportune time to establish this, as not too ...</span></span></sup>.</p>\n<h3>Arguments in favour of the ban</h3>\n<ul>\n<li><strong>Human authorship as a core value:</strong> Wikipedia should remain a project of human-written content; establishing this principle now is timely before too much AI content accumulates.</li>\n<li><strong>Quality and trust:</strong> Preventing AI-generated \"slop\" (<em>AI-Müll</em>) protects Wikipedia's reputation as a reliable, human-authored reference work.</li>\n<li><strong>Window of opportunity:</strong> Proponents argued that acting early is easier than trying to clean up after the fact.</li>\n</ul>\n<h3>Arguments against the ban</h3>\n<ul>\n<li><strong>Detection difficulties:</strong> There are no reliable criteria to identify AI-generated content, risking false positives and unfair accusations against good-faith editors.</li>\n<li><strong>Enforcement gaps:</strong> The proposal lacked detailed procedures for handling violations and repeat offenders.</li>\n<li><strong>Driving use underground:</strong> A blanket ban could push AI use into secrecy rather than eliminating it.</li>\n<li><strong>Excluding beneficial tools:</strong> Future AI-assisted tools (e.g. for translation, copyediting, accessibility) might be caught by an overly broad rule.</li>\n<li><strong>Focus on quality, not method:</strong> Opponents argued the focus should be on whether content meets Wikipedia's standards, not on how it was produced.</li>\n<li><strong>Perception risk:</strong> A rigid ban could make German Wikipedia seem outdated compared to projects that integrate AI thoughtfully.</li>\n<li><strong>Majority threshold dispute:</strong> Some argued such a significant rule change should require a qualified (⅔) majority rather than a simple one.</li>\n</ul>\n<h3>Outcome</h3>\n<p>The Meinungsbild was formally accepted, and the proposal to create the new guideline page <strong>Wikipedia:Künstliche Intelligenz</strong> was also accepted with the required majority. Approximately 350 eligible users voted <sup class=\"trigger\"><a href=\"https://de.wikipedia.org/wiki/Wikipedia:Meinungsbilder/KI-Verbot#:~:text=The%20%27KI-Verbot%27%20%28AI,cast%20their%20votes.\" target=\"_blank\">2</a><span class=\"tooltip top center citation\"><a href=\"https://de.wikipedia.org/wiki/Wikipedia:Meinungsbilder/KI-Verbot#:~:text=The%20%27KI-Verbot%27%20%28AI,cast%20their%20votes.\"><span class=\"source\"><img alt=\"\" src=\"https://p.kagi.com/proxy/favicons?c=lQ30-cxV_E2Vdyqb6hrbuYgId7qu7hecjO7o1T7aqQDFc7XpcLttL1Ryala-GNHJAQ-fjt8WMTZhLN6U1wzeZKYM0S0jk4QmuuY129Xs3iY%3D\"> de.wikipedia.org</span><b>2. Wikipedia:Meinungsbilder/KI-Verbot – Wikipedia</b></a><span class=\"passage\">The 'KI-Verbot' (AI ban) discussion aimed to determine whether the use of Artificial Intelligence, particularly Large Language Models (LLMs), should be largely prohibited in Wikipedia. Arguments for the ban emphasized preserving Wikipedia as a project of human-written content, preventing the influx of AI-generated \"slop\" (AI-Müll), and maintaining Wikipedia's reputation as a beacon of human-authored content. The proponents argued that now is the opportune time to establish this, as not too ...</span></span></sup>.</p>\n<hr>\n<h2>Comparison with English Wikipedia</h2>\n<p>English Wikipedia has taken a more incremental, piecemeal approach rather than a single sweeping vote:</p>\n<table>\n<thead>\n<tr>\n<th>Aspect</th>\n<th>German Wikipedia</th>\n<th>English Wikipedia</th>\n</tr>\n</thead>\n<tbody>\n<tr>\n<td><strong>Approach</strong></td>\n<td>Single binding Meinungsbild creating a dedicated guideline page</td>\n<td>Gradual, distributed policy development across multiple pages</td>\n</tr>\n<tr>\n<td><strong>Text content</strong></td>\n<td>Near-blanket prohibition via new guideline</td>\n<td>No outright ban; editors are responsible for ensuring AI-assisted contributions meet existing policies (verifiability, NOR, etc.) <sup class=\"trigger\"><a href=\"https://en.wikipedia.org/wiki/Wikipedia:AI-generated_content#:~:text=Various%20types%20of,claims%20are%20accurate.\" target=\"_blank\">3</a><span class=\"tooltip top center citation\"><a href=\"https://en.wikipedia.org/wiki/Wikipedia:AI-generated_content#:~:text=Various%20types%20of,claims%20are%20accurate.\"><span class=\"source\"><img alt=\"\" src=\"https://p.kagi.com/proxy/favicons?c=lQ30-cxV_E2Vdyqb6hrbuYgId7qu7hecjO7o1T7aqQDFc7XpcLttL1Ryala-GNHJuWyqBKj8MT4ZBZYnSH4uzuNLNQdbiyz2-gUDYI6hGs0%3D\"> en.wikipedia.org</span><b>3. Wikipedia:AI-generated content</b></a><span class=\"passage\">Various types of computer algorithms, neural networks, machine learning, and approaches sometimes termed artificial intelligence are used to assist humans in generating creative material. Each editor of Wikipedia is responsible for ensuring that everything they submit to Wikipedia meets its policies, guidelines, standards and best practices, and that any and all facts and claims are accurate.</span></span></sup></td>\n</tr>\n<tr>\n<td><strong>Speedy deletion</strong></td>\n<td>Covered under the new guideline</td>\n<td>Since August 2025, suspected AI-generated articles can be nominated for speedy deletion <sup class=\"trigger\"><a href=\"https://en.wikipedia.org/wiki/Artificial_intelligence_in_Wikimedia_projects#:~:text=%5B1%5D%20Some%20applications,for%20speedy%20deletion.\" target=\"_blank\">4</a><span class=\"tooltip top center citation\"><a href=\"https://en.wikipedia.org/wiki/Artificial_intelligence_in_Wikimedia_projects#:~:text=%5B1%5D%20Some%20applications,for%20speedy%20deletion.\"><span class=\"source\"><img alt=\"\" src=\"https://p.kagi.com/proxy/favicons?c=lQ30-cxV_E2Vdyqb6hrbuYgId7qu7hecjO7o1T7aqQDFc7XpcLttL1Ryala-GNHJuWyqBKj8MT4ZBZYnSH4uzuNLNQdbiyz2-gUDYI6hGs0%3D\"> en.wikipedia.org</span><b>4. Artificial intelligence in Wikimedia projects - Wikipedia</b></a><span class=\"passage\">[1] Some applications of artificial intelligence, like using large language models to create new articles from scratch, have been more controversial than others for the Wikipedia community. In August 2025, Wikipedia adopted a policy that allowed editors to nominate suspected AI-generated articles for speedy deletion.</span></span></sup></td>\n</tr>\n<tr>\n<td><strong>Talk pages</strong></td>\n<td>Part of the general guideline</td>\n<td>Specific provision at WP:AITALK allowing striking/collapsing of obviously LLM-generated comments <sup class=\"trigger\"><a href=\"https://en.wikipedia.org/wiki/Wikipedia:WikiProject_AI_Cleanup/Policies#:~:text=Wikipedia%3ATalk%20page%20guidelines,in%20this%20context%3A\" target=\"_blank\">5</a><span class=\"tooltip top center citation\"><a href=\"https://en.wikipedia.org/wiki/Wikipedia:WikiProject_AI_Cleanup/Policies#:~:text=Wikipedia%3ATalk%20page%20guidelines,in%20this%20context%3A\"><span class=\"source\"><img alt=\"\" src=\"https://p.kagi.com/proxy/favicons?c=lQ30-cxV_E2Vdyqb6hrbuYgId7qu7hecjO7o1T7aqQDFc7XpcLttL1Ryala-GNHJuWyqBKj8MT4ZBZYnSH4uzuNLNQdbiyz2-gUDYI6hGs0%3D\"> en.wikipedia.org</span><b>5. Wikipedia:WikiProject AI Cleanup/Policies - Wikipedia</b></a><span class=\"passage\">Wikipedia:Talk page guidelines § LLM-generated (WP:AITALK), a guideline entry allowing the striking or collapsing of comments that are obviously generated by an LLM or similar AI technology The following are not policies or guidelines, but still have some significance in this context:</span></span></sup></td>\n</tr>\n<tr>\n<td><strong>Images</strong></td>\n<td>Included in general AI guideline</td>\n<td>Handled separately: AI-generated images generally barred from mainspace (WP:AIIMAGE) <sup class=\"trigger\"><a href=\"https://en.wikipedia.org/wiki/Wikipedia:AI_image_use#:~:text=Images%20generated%20wholly,notable%20in%20itself.\" target=\"_blank\">7</a><span class=\"tooltip top center citation\"><a href=\"https://en.wikipedia.org/wiki/Wikipedia:AI_image_use#:~:text=Images%20generated%20wholly,notable%20in%20itself.\"><span class=\"source\"><img alt=\"\" src=\"https://p.kagi.com/proxy/favicons?c=lQ30-cxV_E2Vdyqb6hrbuYgId7qu7hecjO7o1T7aqQDFc7XpcLttL1Ryala-GNHJuWyqBKj8MT4ZBZYnSH4uzuNLNQdbiyz2-gUDYI6hGs0%3D\"> en.wikipedia.org</span><b>7. Wikipedia:AI image use</b></a><span class=\"passage\">Images generated wholly by AI should generally not be used to illustrate Wikipedia articles. Exceptions to this rule exist when the image is notable in itself.</span></span></sup><sup class=\"trigger\"><a href=\"https://en.wikipedia.org/wiki/Wikipedia:Image_use_policy#:~:text=AI-generated%20images%E2%80%8B%E2%80%8B%20Most,or%20scientific%20subjects\" target=\"_blank\">8</a><span class=\"tooltip top center citation\"><a href=\"https://en.wikipedia.org/wiki/Wikipedia:Image_use_policy#:~:text=AI-generated%20images%E2%80%8B%E2%80%8B%20Most,or%20scientific%20subjects\"><span class=\"source\"><img alt=\"\" src=\"https://p.kagi.com/proxy/favicons?c=lQ30-cxV_E2Vdyqb6hrbuYgId7qu7hecjO7o1T7aqQDFc7XpcLttL1Ryala-GNHJuWyqBKj8MT4ZBZYnSH4uzuNLNQdbiyz2-gUDYI6hGs0%3D\"> en.wikipedia.org</span><b>8. Wikipedia:Image use policy</b></a><span class=\"passage\">AI-generated images Most images wholly generated by AI should not be used in mainspace, especially for named people and in technical or scientific subjects ...</span></span></sup></td>\n</tr>\n<tr>\n<td><strong>Detection guidance</strong></td>\n<td>Less detailed</td>\n<td>Extensive field guide at WP:Signs of AI writing <sup class=\"trigger\"><a href=\"https://en.wikipedia.org/wiki/Wikipedia:Signs_of_AI_writing#:~:text=Leads%20treating%20Wikipedia,such%20as%20a\" target=\"_blank\">6</a><span class=\"tooltip top center citation\"><a href=\"https://en.wikipedia.org/wiki/Wikipedia:Signs_of_AI_writing#:~:text=Leads%20treating%20Wikipedia,such%20as%20a\"><span class=\"source\"><img alt=\"\" src=\"https://p.kagi.com/proxy/favicons?c=lQ30-cxV_E2Vdyqb6hrbuYgId7qu7hecjO7o1T7aqQDFc7XpcLttL1Ryala-GNHJuWyqBKj8MT4ZBZYnSH4uzuNLNQdbiyz2-gUDYI6hGs0%3D\"> en.wikipedia.org</span><b>6. Wikipedia:Signs of AI writing</b></a><span class=\"passage\">Leads treating Wikipedia lists or broad article titles as proper nouns. In AI-generated articles about topics with a title that is not a proper name, such as a ...</span></span></sup></td>\n</tr>\n</tbody>\n</table>\n<h3>Useful takeaways for English Wikipedia</h3>\n<ol>\n<li>\n<p><strong>Centralisation has merits.</strong> German Wikipedia's single guideline page avoids the fragmentation English Wikipedia currently has (rules scattered across WP:AIGEN, WP:AIIMAGE, WP:AITALK, CSD criteria, etc.). A consolidated \"hub\" page — even if it primarily cross-references existing policies — could improve discoverability and coherence. English Wikipedia's <strong>WP:Artificial intelligence</strong> serves this role to some extent <sup class=\"trigger\"><a href=\"https://en.wikipedia.org/wiki/Wikipedia:Artificial_intelligence#:~:text=Artificial%20intelligence%20%28AI%29,to%20exercise%20more\" target=\"_blank\">1</a><span class=\"tooltip top center citation\"><a href=\"https://en.wikipedia.org/wiki/Wikipedia:Artificial_intelligence#:~:text=Artificial%20intelligence%20%28AI%29,to%20exercise%20more\"><span class=\"source\"><img alt=\"\" src=\"https://p.kagi.com/proxy/favicons?c=lQ30-cxV_E2Vdyqb6hrbuYgId7qu7hecjO7o1T7aqQDFc7XpcLttL1Ryala-GNHJuWyqBKj8MT4ZBZYnSH4uzuNLNQdbiyz2-gUDYI6hGs0%3D\"> en.wikipedia.org</span><b>1. Wikipedia:Artificial intelligence - Wikipedia</b></a><span class=\"passage\">Artificial intelligence (AI) is used on a number of Wikipedia and Wikimedia projects. This may be directly involved with creation of text content, or in support roles related to evaluating article quality, adding metadata, or generating images. As with any machine-generated content, care must be used when employing AI at scale or in applying it where the community consensus is to exercise more ...</span></span></sup>, but it is an information page, not a guideline.</p>\n</li>\n<li>\n<p><strong>The \"quality not method\" counterargument resonates.</strong> English Wikipedia's existing framework already leans this way — holding editors accountable for what they submit regardless of how they produced it <sup class=\"trigger\"><a href=\"https://en.wikipedia.org/wiki/Wikipedia:AI-generated_content#:~:text=Various%20types%20of,claims%20are%20accurate.\" target=\"_blank\">3</a><span class=\"tooltip top center citation\"><a href=\"https://en.wikipedia.org/wiki/Wikipedia:AI-generated_content#:~:text=Various%20types%20of,claims%20are%20accurate.\"><span class=\"source\"><img alt=\"\" src=\"https://p.kagi.com/proxy/favicons?c=lQ30-cxV_E2Vdyqb6hrbuYgId7qu7hecjO7o1T7aqQDFc7XpcLttL1Ryala-GNHJuWyqBKj8MT4ZBZYnSH4uzuNLNQdbiyz2-gUDYI6hGs0%3D\"> en.wikipedia.org</span><b>3. Wikipedia:AI-generated content</b></a><span class=\"passage\">Various types of computer algorithms, neural networks, machine learning, and approaches sometimes termed artificial intelligence are used to assist humans in generating creative material. Each editor of Wikipedia is responsible for ensuring that everything they submit to Wikipedia meets its policies, guidelines, standards and best practices, and that any and all facts and claims are accurate.</span></span></sup>. The German debate shows this position can coexist with strong anti-AI-slop measures, but that a pure quality-based approach can feel insufficient to editors who see AI use as a values question.</p>\n</li>\n<li>\n<p><strong>Enforcement and detection remain the hard problem.</strong> Both communities acknowledge that reliable detection is elusive. The German debate's concern about false positives and driving use underground is directly relevant to any English Wikipedia proposals to strengthen sanctions.</p>\n</li>\n<li>\n<p><strong>Scope creep risk.</strong> The German opposition's worry about excluding beneficial AI tools (translation aids, accessibility features, anti-vandalism bots) is worth bearing in mind. English Wikipedia already uses ML-based tools extensively (ORES, ClueBot NG, etc.) <sup class=\"trigger\"><a href=\"https://en.wikipedia.org/wiki/Wikipedia:Artificial_intelligence#:~:text=Artificial%20intelligence%20%28AI%29,to%20exercise%20more\" target=\"_blank\">1</a><span class=\"tooltip top center citation\"><a href=\"https://en.wikipedia.org/wiki/Wikipedia:Artificial_intelligence#:~:text=Artificial%20intelligence%20%28AI%29,to%20exercise%20more\"><span class=\"source\"><img alt=\"\" src=\"https://p.kagi.com/proxy/favicons?c=lQ30-cxV_E2Vdyqb6hrbuYgId7qu7hecjO7o1T7aqQDFc7XpcLttL1Ryala-GNHJuWyqBKj8MT4ZBZYnSH4uzuNLNQdbiyz2-gUDYI6hGs0%3D\"> en.wikipedia.org</span><b>1. Wikipedia:Artificial intelligence - Wikipedia</b></a><span class=\"passage\">Artificial intelligence (AI) is used on a number of Wikipedia and Wikimedia projects. This may be directly involved with creation of text content, or in support roles related to evaluating article quality, adding metadata, or generating images. As with any machine-generated content, care must be used when employing AI at scale or in applying it where the community consensus is to exercise more ...</span></span></sup><sup class=\"trigger\"><a href=\"https://en.wikipedia.org/wiki/Artificial_intelligence_in_Wikimedia_projects#:~:text=%5B1%5D%20Some%20applications,for%20speedy%20deletion.\" target=\"_blank\">4</a><span class=\"tooltip top center citation\"><a href=\"https://en.wikipedia.org/wiki/Artificial_intelligence_in_Wikimedia_projects#:~:text=%5B1%5D%20Some%20applications,for%20speedy%20deletion.\"><span class=\"source\"><img alt=\"\" src=\"https://p.kagi.com/proxy/favicons?c=lQ30-cxV_E2Vdyqb6hrbuYgId7qu7hecjO7o1T7aqQDFc7XpcLttL1Ryala-GNHJuWyqBKj8MT4ZBZYnSH4uzuNLNQdbiyz2-gUDYI6hGs0%3D\"> en.wikipedia.org</span><b>4. Artificial intelligence in Wikimedia projects - Wikipedia</b></a><span class=\"passage\">[1] Some applications of artificial intelligence, like using large language models to create new articles from scratch, have been more controversial than others for the Wikipedia community. In August 2025, Wikipedia adopted a policy that allowed editors to nominate suspected AI-generated articles for speedy deletion.</span></span></sup>, so any future tightening needs careful scoping to avoid unintended consequences.</p>\n</li>\n<li>\n<p><strong>Community mandate matters.</strong> The Meinungsbild format — with its formal acceptance threshold and large turnout (350 voters) — gave the result clear legitimacy. English Wikipedia's piecemeal approach risks individual guidelines being challenged as lacking sufficient consensus; a broad RfC or similar exercise could provide a stronger foundation.</p>\n</li>\n</ol>","md":"<details><summary>Thinking</summary>\n\nThe user wants a summary of the German Wikipedia \"KI-Verbot\" (AI ban) Meinungsbild, plus takeaways for English Wikipedia. Let me synthesize the information I have.\n\n</details>\n\n## German Wikipedia: KI-Verbot Meinungsbild\n\n### Overview\n\nThe *Meinungsbild* (a formal community opinion poll, roughly analogous to an English Wikipedia RfC but with binding force) proposed to largely prohibit the use of AI — particularly large language models — in German Wikipedia, and to create a new guideline page, **Wikipedia:Künstliche Intelligenz**, codifying rules on AI use [^2].\n\n### Arguments in favour of the ban\n\n- **Human authorship as a core value:** Wikipedia should remain a project of human-written content; establishing this principle now is timely before too much AI content accumulates.\n- **Quality and trust:** Preventing AI-generated \"slop\" (*AI-Müll*) protects Wikipedia's reputation as a reliable, human-authored reference work.\n- **Window of opportunity:** Proponents argued that acting early is easier than trying to clean up after the fact.\n\n### Arguments against the ban\n\n- **Detection difficulties:** There are no reliable criteria to identify AI-generated content, risking false positives and unfair accusations against good-faith editors.\n- **Enforcement gaps:** The proposal lacked detailed procedures for handling violations and repeat offenders.\n- **Driving use underground:** A blanket ban could push AI use into secrecy rather than eliminating it.\n- **Excluding beneficial tools:** Future AI-assisted tools (e.g. for translation, copyediting, accessibility) might be caught by an overly broad rule.\n- **Focus on quality, not method:** Opponents argued the focus should be on whether content meets Wikipedia's standards, not on how it was produced.\n- **Perception risk:** A rigid ban could make German Wikipedia seem outdated compared to projects that integrate AI thoughtfully.\n- **Majority threshold dispute:** Some argued such a significant rule change should require a qualified (⅔) majority rather than a simple one.\n\n### Outcome\n\nThe Meinungsbild was formally accepted, and the proposal to create the new guideline page **Wikipedia:Künstliche Intelligenz** was also accepted with the required majority. Approximately 350 eligible users voted [^2].\n\n---\n\n## Comparison with English Wikipedia\n\nEnglish Wikipedia has taken a more incremental, piecemeal approach rather than a single sweeping vote:\n\n| Aspect | German Wikipedia | English Wikipedia |\n|---|---|---|\n| **Approach** | Single binding Meinungsbild creating a dedicated guideline page | Gradual, distributed policy development across multiple pages |\n| **Text content** | Near-blanket prohibition via new guideline | No outright ban; editors are responsible for ensuring AI-assisted contributions meet existing policies (verifiability, NOR, etc.) [^3] |\n| **Speedy deletion** | Covered under the new guideline | Since August 2025, suspected AI-generated articles can be nominated for speedy deletion [^4] |\n| **Talk pages** | Part of the general guideline | Specific provision at WP:AITALK allowing striking/collapsing of obviously LLM-generated comments [^5] |\n| **Images** | Included in general AI guideline | Handled separately: AI-generated images generally barred from mainspace (WP:AIIMAGE) [^7][^8] |\n| **Detection guidance** | Less detailed | Extensive field guide at WP:Signs of AI writing [^6] |\n\n### Useful takeaways for English Wikipedia\n\n1. **Centralisation has merits.** German Wikipedia's single guideline page avoids the fragmentation English Wikipedia currently has (rules scattered across WP:AIGEN, WP:AIIMAGE, WP:AITALK, CSD criteria, etc.). A consolidated \"hub\" page — even if it primarily cross-references existing policies — could improve discoverability and coherence. English Wikipedia's **WP:Artificial intelligence** serves this role to some extent [^1], but it is an information page, not a guideline.\n\n2. **The \"quality not method\" counterargument resonates.** English Wikipedia's existing framework already leans this way — holding editors accountable for what they submit regardless of how they produced it [^3]. The German debate shows this position can coexist with strong anti-AI-slop measures, but that a pure quality-based approach can feel insufficient to editors who see AI use as a values question.\n\n3. **Enforcement and detection remain the hard problem.** Both communities acknowledge that reliable detection is elusive. The German debate's concern about false positives and driving use underground is directly relevant to any English Wikipedia proposals to strengthen sanctions.\n\n4. **Scope creep risk.** The German opposition's worry about excluding beneficial AI tools (translation aids, accessibility features, anti-vandalism bots) is worth bearing in mind. English Wikipedia already uses ML-based tools extensively (ORES, ClueBot NG, etc.) [^1][^4], so any future tightening needs careful scoping to avoid unintended consequences.\n\n5. **Community mandate matters.** The Meinungsbild format — with its formal acceptance threshold and large turnout (350 voters) — gave the result clear legitimacy. English Wikipedia's piecemeal approach risks individual guidelines being challenged as lacking sufficient consensus; a broad RfC or similar exercise could provide a stronger foundation.","profile":{"id":"1b6d99f4-1649-48c8-8ee3-d6774318f814","name":"wikimedia-policy","accessible":true,"model":"claude-4-6-opus-thinking","deprecate":false,"retired":false,"successor_model":null,"model_name":"Claude 4.6 Opus (reasoning)","model_provider":"anthropic","model_input_limit":1000000,"recommended":true,"model_info":"\n\n\n\n<div class=\"heading\">\n <div class=\"left\">\n \n \n \n <svg class=\"model-icon icon-xs\" data-model=\"claude-4-6-opus-thinking\">\n <use href=\"#icon-sparkles\"/>\n </svg>\n <h3>\n \n <span class=\"model-name\">wikimedia-policy</span>\n </h3>\n </div>\n\n \n <a href=\"/settings?p=custom_assistant&id=1b6d99f4-1649-48c8-8ee3-d6774318f814\">\n <svg class=\"icon-sm edit-icon\" aria-hidden=\"true\">\n <use href=\"#icon-pencil\" />\n </svg>\n </a>\n \n</div>\n\n\n<hr />\n<ul>\n \n <li>\n <h4>\n Model\n </h4>\n <span class=\"flex align-center gap-8\">\n <svg class=\"icon-xs icon\">\n <use href=\"#icon-anthropic\" />\n </svg>\n <span class=\"model-name\">Claude 4.6 Opus (reasoning)</span>\n </span>\n </li>\n <li>\n <h4>\n Web access\n </h4>\n <span>On</span>\n </li>\n <li>\n <h4>\n Lenses\n </h4>\n <span></span>\n </li>\n \n</ul>\n\n\n","model_provider_name":"Anthropic","internet_access":true,"personalizations":false,"lens":{"id":21245,"name":null},"shortcut":null,"is_default_profile":false},"metadata":"<li>\n <span class=\"attribute\">Model</span>\n <span class=\"value\"><i class=\"icon-xs mr-4\"><svg><use href=\"#icon-anthropic\"></use></svg></i>Claude 4.6 Opus (reasoning)</span>\n</li>\n<li>\n <span class=\"attribute\">Version</span>\n <span class=\"value\">claude-opus-4-6</span>\n</li>\n<li>\n <span class=\"attribute\">Speed (tok/s)</span>\n <span class=\"value\">43</span>\n</li>\n<li>\n <span class=\"attribute\">Tokens</span>\n <span class=\"value\">5,043</span>\n</li>\n<li>\n <span class=\"attribute\">Cost / Total ($)</span>\n <span class=\"value\">0.07 / 0.07</span>\n</li>\n<li>\n <span class=\"attribute\">End to end time (s)</span>\n <span class=\"value\">34.59</span>\n</li>\n<li>\n <span class=\"attribute\">Submitted</span>\n <span class=\"value\">2026-02-19 18:02</span>\n</li>\n\n","documents":[]},{"id":"9be042e3-24c3-4b7c-add6-d0c7fd59e7fd","thread_id":"674061bf-35fc-496d-8b54-17b25db50fd7","created_at":"2026-02-19T23:11:19Z","branch_list":["076941ae-ab29-4129-8508-fc63755bb3e7"],"state":"done","prompt":"Does German Wikipedia have the same problem with quantity that has been observed on English Wikipedia, where the volume of low-quality LLM-generated material far exceeds the capacity of experienced editors to review or clean it up?","references_html":"<ol data-ref-list>\n \n <li>\n <a href=\"https://de.wikipedia.org/wiki/Wikipedia:WikiProjekt_KI_und_Wikipedia/Wikiversum\" >Wikipedia:WikiProjekt KI und Wikipedia/Wikiversum – Wikipedia</a>\n <span class=\"__domain-name\">de.wikipedia.org</span>\n <div class=\"trigger\" tabindex=\"0\">\n <span class=\"contribution\" style=\"background: linear-gradient(to right, var(--primary-70) 78%, transparent 78%)\">78%</span>\n <div class=\"tooltip top end tooltip_base tw-auto\">\n This percentage indicates how much a source has contributed to the answer.\n </div>\n </div>\n </li>\n \n <li>\n <a href=\"https://de.wikipedia.org/wiki/Wikipedia:WikiProjekt_KI_und_Wikipedia/Regeln\" >Wikipedia:WikiProjekt KI und Wikipedia/Regeln – Wikipedia</a>\n <span class=\"__domain-name\">de.wikipedia.org</span>\n <div class=\"trigger\" tabindex=\"0\">\n <span class=\"contribution\" style=\"background: linear-gradient(to right, var(--primary-70) 22%, transparent 22%)\">22%</span>\n <div class=\"tooltip top end tooltip_base tw-auto\">\n This percentage indicates how much a source has contributed to the answer.\n </div>\n </div>\n </li>\n \n</ol>\n","references_md":"[^1]: [Wikipedia:WikiProjekt KI und Wikipedia/Wikiversum – Wikipedia](https://de.wikipedia.org/wiki/Wikipedia:WikiProjekt_KI_und_Wikipedia/Wikiversum) (78%)\n[^2]: [Wikipedia:WikiProjekt KI und Wikipedia/Regeln – Wikipedia](https://de.wikipedia.org/wiki/Wikipedia:WikiProjekt_KI_und_Wikipedia/Regeln) (22%)\n","reply":"<details><summary>Searched from de.wikipedia.org with Kagi<p></p>\n<ul>\n<li>German Wikipedia KI-generierte Inhalte Menge Überforderung Redakteure Kapazität</li>\n<li>German Wikipedia AI content volume editor capacity</li>\n<li>Wikipedia:Künstliche Intelligenz German Wikipedia discussion</li></ul></summary><br>\nSources (30):<br>\n<strong>Wikipedia:Technik/Werkstatt</strong><br>\n<a href=\"https://de.wikipedia.org/wiki/Wikipedia:Technik/Werkstatt\" target=\"_blank\">https://de.wikipedia.org/wiki/Wikipedia:Technik/Werkstatt</a><br>\nDu hast einen mutmaßlichen Software-Fehler entdeckt, magst das aber nicht auf Phabricator melden? Wir prüfen gern, ob dies ein weltweiter Fehler ist oder ein in ...\n\n<p><strong>Wikipedia:Anzeichen für KI-generierte Inhalte – Wikipedia</strong><br>\n<a href=\"https://de.wikipedia.org/wiki/Wikipedia:Anzeichen_f%C3%BCr_KI-generierte_Inhalte\" target=\"_blank\">https://de.wikipedia.org/wiki/Wikipedia:Anzeichen_f%C3%BCr_KI-generierte_Inhalte</a><br>\n Die Inhalte sind noch nicht mit dem restlichen WikiProjekt KI und Wikipedia abgestimmt. Insbesondere sollten Redundanzen mit Wikipedia:WikiProjekt KI und Wikipedia/Erkennung ungeprüfter KI-Einsatz abgearbeitet und die Seiten zusammengeführt werden. Für Weiteres siehe die Diskussionsseite.</p>\n<p><strong>Wikipedia:Café/Archiv 2023 03</strong><br>\n<a href=\"https://de.wikipedia.org/wiki/Wikipedia:Caf%C3%A9/Archiv_2023_03\" target=\"_blank\">https://de.wikipedia.org/wiki/Wikipedia:Caf%C3%A9/Archiv_2023_03</a><br>\n In so ziemlich jeder schlechten KI-SF dreht die KI durch und wird zur Gefahr oder führt zu Verdummung, bis Kirk sie umboxt oder Spock sie austrickst. Die ...</p>\n<p><strong>Wikipedia:WikiProjekt KI und Wikipedia/Wikiversum – Wikipedia</strong><br>\n<a href=\"https://de.wikipedia.org/wiki/Wikipedia:WikiProjekt_KI_und_Wikipedia/Wikiversum\" target=\"_blank\">https://de.wikipedia.org/wiki/Wikipedia:WikiProjekt_KI_und_Wikipedia/Wikiversum</a><br>\n Unbefriedigend beantwortet bleibt, ob die generative KI die Qualitätskontrollmechanismen von Wikipedia unterläuft - oder ob diese LLM-generierte Artikel neben dem bestehenden nie endenden Strom von vom Menschen erzeugtem Vandalismus, Hoaxes oder Artikeln mit fehlenden oder irreführenden Referenzen gut erkannt werden.</p>\n<p><strong>Wikipedia:Auskunft/Archiv/2007/Sep</strong><br>\n<a href=\"https://de.wikipedia.org/wiki/Wikipedia:Auskunft/Archiv/2007/Sep\" target=\"_blank\">https://de.wikipedia.org/wiki/Wikipedia:Auskunft/Archiv/2007/Sep</a><br>\n Dies ist ein Archiv der Seite Wikipedia:Auskunft. Möchtest du in einer neuen Diskussion auf dieses Archiv verweisen? Nutze folgenden Link und ersetze ______ ...</p>\n<p><strong>Wikipedia:WikiProjekt KI und Wikipedia – Wikipedia</strong><br>\n<a href=\"https://de.wikipedia.org/wiki/Wikipedia:WikiProjekt_KI_und_Wikipedia\" target=\"_blank\">https://de.wikipedia.org/wiki/Wikipedia:WikiProjekt_KI_und_Wikipedia</a><br>\n Erproben von KI-Werkzeugen als Hilfsmittel und Bericht über Erfahrungen, um Möglichkeiten und Grenzen zu erkennen. Da das notwendig ist, um den Anschluss nicht ganz zu verlieren, kann das Verbot bzw. die Unerwünschtheit für die Darstellung von KI-erzeugten Inhalten zu Demonstrations- und Testzwecken innerhalb des Projektes nicht gelten.</p>\n<p><strong>Wikipedia:Café/Archiv 2024 07</strong><br>\n<a href=\"https://de.wikipedia.org/wiki/Wikipedia:Caf%C3%A9/Archiv_2024_07\" target=\"_blank\">https://de.wikipedia.org/wiki/Wikipedia:Caf%C3%A9/Archiv_2024_07</a><br>\n Diese Seite ist ein Archiv abgeschlossener Diskussionen. Ihr Inhalt sollte daher nicht mehr verändert werden. Benutze bitte die aktuelle Projektseite, ...</p>\n<p><strong>Wikipedia:WikiProjekt KI und Wikipedia/KI-generierte Artikel – Wikipedia</strong><br>\n<a href=\"https://de.wikipedia.org/wiki/Wikipedia:WikiProjekt_KI_und_Wikipedia/KI-generierte_Artikel\" target=\"_blank\">https://de.wikipedia.org/wiki/Wikipedia:WikiProjekt_KI_und_Wikipedia/KI-generierte_Artikel</a><br>\n KI für Zusammenfassungen: siehe Reading/Web/Content Discovery Experiment auf mediaWiki insbesondere Experiment #4 (in der englischsprachigen WP sind Zusammenfassungen üblich, die über unsere Einleitung hinaus gehen).</p>\n<p><strong>Wikipedia:Café/Archiv 2023 09</strong><br>\n<a href=\"https://de.wikipedia.org/wiki/Wikipedia:Caf%C3%A9/Archiv_2023_09\" target=\"_blank\">https://de.wikipedia.org/wiki/Wikipedia:Caf%C3%A9/Archiv_2023_09</a><br>\n Diese Seite ist ein Archiv abgeschlossener Diskussionen. Ihr Inhalt sollte daher nicht mehr verändert werden. Benutze bitte die aktuelle Projektseite, ...</p>\n<p><strong>Wikipedia:WikiProjekt KI und Wikipedia/Regeln – Wikipedia</strong><br>\n<a href=\"https://de.wikipedia.org/wiki/Wikipedia:WikiProjekt_KI_und_Wikipedia/Regeln\" target=\"_blank\">https://de.wikipedia.org/wiki/Wikipedia:WikiProjekt_KI_und_Wikipedia/Regeln</a><br>\n Ziel der Sensibilisierung, wenn KI als Hilfsmittel verwendet wird, ist der Hinweis auf Fehlermöglichkeiten und die Verhinderung, das AI Slop (KI Müll) Eingang in die Wikipedia findet. Begriffe: Ein Textteil sind zusammengehörige inhaltliche Aussagen, die belegt (REF) sein müssen und eine Quelle, ein Werk, richtig darstellen. Ein Artikel besteht in der Regel aus mehreren Textteilen. Humans ...</p>\n<p><strong>Wikipedia:Wikidata/Newsletter</strong><br>\n<a href=\"https://de.wikipedia.org/wiki/Wikipedia:Wikidata/Newsletter\" target=\"_blank\">https://de.wikipedia.org/wiki/Wikipedia:Wikidata/Newsletter</a><br>\n Empowering Open-Source Generative AI by Integrating the Wikidata knowledge graph --> Generative AI ... Water bottle volume (Volume of the water bottle); Is it ...</p>\n<p><strong>Wikipedia:WikiProjekt KI und Wikipedia – Wikipedia</strong><br>\n<a href=\"https://de.wikipedia.org/wiki/Wikipedia:WikiProjekt_KI_und_Wikipedia\" target=\"_blank\">https://de.wikipedia.org/wiki/Wikipedia:WikiProjekt_KI_und_Wikipedia</a><br>\n Interesse Stichwortsuche auf allen Projektseiten (ohne Diskussionsseiten) Visual Editor Auswirkungen generativer KI auf Arbeitsweise, Qualitätssicherung und Regelwerk der Wikipedia heute und morgen, Risiken und Chancen der KI für Wikipedia systematisch bestimmen.</p>\n<p><strong>Wikipedia:Wikidata/Newsletter/Archiv/2025</strong><br>\n<a href=\"https://de.wikipedia.org/wiki/Wikipedia:Wikidata/Newsletter/Archiv/2025\" target=\"_blank\">https://de.wikipedia.org/wiki/Wikipedia:Wikidata/Newsletter/Archiv/2025</a><br>\n EduWiki Conference - Using AI on Wikipedia and Wikidata Editing for Health Education · Using Duplicity: Link Wikipedia articles to Wikidata Items · How to Get ...</p>\n<p><strong>de.wikipedia.org</strong><br>\n<a href=\"https://de.wikipedia.org/wiki/Wijzigingenin%27volume%27\" target=\"_blank\">https://de.wikipedia.org/wiki/Wijzigingenin%27volume%27</a><br>\n de.wikipedia.org</p>\n<p><strong>Liste von Abkürzungen (Computer) - Wikipedia</strong><br>\n<a href=\"https://de.wikipedia.org/wiki/Liste_von_Abk%C3%BCrzungen_(Computer\" target=\"_blank\">https://de.wikipedia.org/wiki/Liste_von_Abk%C3%BCrzungen_(Computer)</a>)<br>\n Dies ist eine Liste technischer Abkürzungen, die im IT-Bereich verwendet werden. Inhaltsverzeichnis A B C D E F G H I J K L M N O P Q R S T U V W X Y Z ...</p>\n<p><strong>Künstliche Intelligenz – Wikipedia</strong><br>\n<a href=\"https://de.wikipedia.org/wiki/K%C3%BCnstliche_Intelligenz\" target=\"_blank\">https://de.wikipedia.org/wiki/K%C3%BCnstliche_Intelligenz</a><br>\n Künstliche Intelligenz (KI), englisch artificial intelligence, daher auch artifizielle Intelligenz (AI), bezeichnet im weitesten Sinne computerbasierte Systeme, die ihre (virtuelle oder reale) Umgebung analysieren können, um daraus relevante Informationen zu abstrahieren, welche sie nutzen, um Entscheidungen zu treffen, die ihre Chance ...</p>\n<p><strong>Wikipedia:Wikidata/Newsletter/Archiv/2024</strong><br>\n<a href=\"https://de.wikipedia.org/wiki/Wikipedia:Wikidata/Newsletter/Archiv/2024\" target=\"_blank\">https://de.wikipedia.org/wiki/Wikipedia:Wikidata/Newsletter/Archiv/2024</a><br>\n Diese Seite ist ein Archiv abgeschlossener Diskussionen. Ihr Inhalt sollte daher nicht mehr verändert werden. Benutze bitte die aktuelle Projektseite, ...</p>\n<p><strong>Wikipedia:WikiProjekt KI und Wikipedia/KI und Bilder – Wikipedia</strong><br>\n<a href=\"https://de.wikipedia.org/wiki/Wikipedia:WikiProjekt_KI_und_Wikipedia/KI_und_Bilder\" target=\"_blank\">https://de.wikipedia.org/wiki/Wikipedia:WikiProjekt_KI_und_Wikipedia/KI_und_Bilder</a><br>\n AI-generierte Bilder oder andere Medien auf Commons siehe c:Category:AI-generated media COMMONS: Umgang mit KI-generierten Medien: Per the Commons project scope, only media that are realistically useful for an educational purpose should be hosted on Commons. Just because an AI image is interesting, pretty, or looks like a work of art, that doesn't mean that it is necessarily within the scope ...</p>\n<p><strong>Wikipedia:Wikidata/Newsletter/Archiv/2023</strong><br>\n<a href=\"https://de.wikipedia.org/wiki/Wikipedia:Wikidata/Newsletter/Archiv/2023\" target=\"_blank\">https://de.wikipedia.org/wiki/Wikipedia:Wikidata/Newsletter/Archiv/2023</a><br>\n Diese Seite ist ein Archiv abgeschlossener Diskussionen. Ihr Inhalt sollte daher nicht mehr verändert werden. Benutze bitte die aktuelle Projektseite, ...</p>\n<p><strong>Wikipedia:Technik/Text/Edit/VisualEditor – Wikipedia</strong><br>\n<a href=\"https://de.wikipedia.org/wiki/Wikipedia:Technik/Text/Edit/VisualEditor\" target=\"_blank\">https://de.wikipedia.org/wiki/Wikipedia:Technik/Text/Edit/VisualEditor</a><br>\n Der VisualEditor wird im Rahmen eines Projekts der Wikimedia Foundation entwickelt. Ziel dieses Projekts ist die Erstellung einer WYSIWYG - Bearbeitungsoberfläche für MediaWiki, die es den Benutzern ermöglicht, Änderungen an Artikeltexten vorzunehmen, ohne die zugrundeliegende Auszeichnungssprache von MediaWiki zu verwenden. Die Verfügbarkeit dieser „ Online-Textverarbeitung “ soll ...</p>\n<p><strong>Artificial General Intelligence - Wikipedia</strong><br>\n<a href=\"https://de.wikipedia.org/wiki/Artificial_General_Intelligence\" target=\"_blank\">https://de.wikipedia.org/wiki/Artificial_General_Intelligence</a><br>\n Artificial general intelligence (AGI), deutsch künstliche allgemeine Intelligenz, ist die Intelligenz eines hypothetischen Computerprogramms, welches die Fähigkeit besitzt, jede intellektuelle Aufgabe zu verstehen oder zu lernen, die ein Mensch ausführen kann.</p>\n<p><strong>Künstliche Intelligenz – Wikipedia</strong><br>\n<a href=\"https://de.wikipedia.org/wiki/K%C3%BCnstliche_Intelligenz\" target=\"_blank\">https://de.wikipedia.org/wiki/K%C3%BCnstliche_Intelligenz</a><br>\n Künstliche Intelligenz (KI), englisch artificial intelligence, daher auch artifizielle Intelligenz (AI), bezeichnet im weitesten Sinne computerbasierte Systeme, die ihre (virtuelle oder reale) Umgebung analysieren können, um daraus relevante Informationen zu abstrahieren, welche sie nutzen, um Entscheidungen zu treffen, die ihre Chance ...</p>\n<p><strong>Existenzielles Risiko durch künstliche Intelligenz - Wikipedia</strong><br>\n<a href=\"https://de.wikipedia.org/wiki/Existenzielles_Risiko_durch_k%C3%BCnstliche_Intelligenz\" target=\"_blank\">https://de.wikipedia.org/wiki/Existenzielles_Risiko_durch_k%C3%BCnstliche_Intelligenz</a><br>\n Wie plausibel eine existenzielle Katastrophe durch Künstliche Intelligenz (KI) ist, wird breit diskutiert und hängt zum Teil davon ab, ob AGI oder ...</p>\n<p><strong>Wikipedia:Künstliche Intelligenz – Wikipedia</strong><br>\n<a href=\"https://de.wikipedia.org/wiki/Wikipedia:K%C3%BCnstliche_Intelligenz\" target=\"_blank\">https://de.wikipedia.org/wiki/Wikipedia:K%C3%BCnstliche_Intelligenz</a><br>\n Außerhalb des Themenbereichs Künstliche Intelligenz dürfen KI-generierte Bilder im Artikelnamensraum nur in Ausnahmefällen, nach Diskussion und Konsens für jedes einzelne Bild, verwendet werden. Es ist nicht erlaubt, ein KI-generiertes Bild zu verwenden, ohne vorher auf der zugehörigen Diskussionsseite einen Konsens dafür gefunden zu haben.</p>\n<p><strong>Explainable Artificial Intelligence - Wikipedia</strong><br>\n<a href=\"https://de.wikipedia.org/wiki/Explainable_Artificial_Intelligence\" target=\"_blank\">https://de.wikipedia.org/wiki/Explainable_Artificial_Intelligence</a><br>\n Explainable Artificial Intelligence · Definition · Geschichte · Methoden · Anwendungsbeispiele · Internationaler Austausch über die Thematik und Forderungen.</p>\n<p><strong>Geschichte der künstlichen Intelligenz – Wikipedia</strong><br>\n<a href=\"https://de.wikipedia.org/wiki/Geschichte_der_k%C3%BCnstlichen_Intelligenz\" target=\"_blank\">https://de.wikipedia.org/wiki/Geschichte_der_k%C3%BCnstlichen_Intelligenz</a><br>\n Theoretische Grundlagen: Konzepte zur Formalisierung des Denkens Künstliche Intelligenz basiert auf der Annahme, dass der Prozess des menschlichen Denkens formalisiert werden kann. Das Studium des mechanischen – oder formalen – Denkens hat eine lange Geschichte.</p>\n<p><strong>Sora (künstliche Intelligenz) - Wikipedia</strong><br>\n<a href=\"https://de.wikipedia.org/wiki/Sora_(k%C3%BCnstliche_Intelligenz\" target=\"_blank\">https://de.wikipedia.org/wiki/Sora_(k%C3%BCnstliche_Intelligenz)</a>)<br>\n Sora ist eine von OpenAI entwickelte generative künstliche Intelligenz, die auf die Erstellung von Videos aus Texteingaben spezialisiert ist ...</p>\n<p><strong>Wikipedia:WikiProjekt KI und Wikipedia/KI verstehen – Wikipedia</strong><br>\n<a href=\"https://de.wikipedia.org/wiki/Wikipedia:WikiProjekt_KI_und_Wikipedia/KI_verstehen\" target=\"_blank\">https://de.wikipedia.org/wiki/Wikipedia:WikiProjekt_KI_und_Wikipedia/KI_verstehen</a><br>\n Was sollte oder müsste man wissen, um das Thema \"KI heute\" zu verstehen? In bisherigen Diskussionen redet man oft aneinander vorbei, weil nicht vom Gleichen. Die Geschichte der künstlichen Intelligenz: eine Zeitleiste ComputerWeekly 5. August 2025 neues Buch Frank Schmiedchen u.a. (Hrsg.):Künstliche Intelligenz und Wir Stand, Nutzung und Herausforderungen der KI Springer 2026 Klappentext ...</p>\n<p><strong>Intelligenz - Wikipedia</strong><br>\n<a href=\"https://de.wikipedia.org/wiki/Intelligenz\" target=\"_blank\">https://de.wikipedia.org/wiki/Intelligenz</a><br>\n Mit Intelligenz und Intelligenztests befassen sich speziell die Allgemeine Psychologie, die Differentielle Psychologie und die Neuropsychologie. Die Erforschung ...</p>\n<p><strong>Wikipedia:WikiProjekt KI und Wikipedia/KI und Bilder</strong><br>\n<a href=\"https://de.wikipedia.org/wiki/Wikipedia:WikiProjekt_KI_und_Wikipedia/KI_und_Bilder\" target=\"_blank\">https://de.wikipedia.org/wiki/Wikipedia:WikiProjekt_KI_und_Wikipedia/KI_und_Bilder</a><br>\n Realistisch wirkende Videos ohne Ton erzeugt eine neue generative Künstliche Intelligenz (KI) aus dem Hause OpenAI. Der Dienst heißt Sora und erstellt auf ...</p>\n</details><details><summary>Thinking</summary>\n<p>The user is asking whether German Wikipedia faces the same scale problem as English Wikipedia with AI-generated content overwhelming reviewer capacity. Let me work with what I know and what the research has found.</p>\n<p>Key contextual differences between the two projects:</p>\n<ol>\n<li>\n<p><strong>Scale</strong>: English Wikipedia is far larger (~6.9 million articles vs ~2.9 million on German Wikipedia) and receives far more new article submissions, including from the Global South where LLM-generated content in English is a particular vector.</p>\n</li>\n<li>\n<p><strong>New page patrol</strong>: English Wikipedia has a massive backlog problem with new page patrol (NPP), which is where most AI-generated articles first appear. German Wikipedia has a different new article review process.</p>\n</li>\n<li>\n<p><strong>Language barrier</strong>: English is the lingua franca and the primary output language for most LLMs, making English Wikipedia a more attractive target. German Wikipedia benefits from some insulation through language, though LLMs are increasingly competent at generating German text.</p>\n</li>\n<li>\n<p><strong>The Meinungsbild itself</strong>: The near-blanket ban that German Wikipedia passed indicates the community saw a problem worth addressing, yet the research suggests whether AI content is actually overwhelming quality control remains an open question.</p>\n</li>\n</ol>\n</details>\n<h2>AI content volume: German vs English Wikipedia</h2>\n<h3>The short answer</h3>\n<p>German Wikipedia faces the same <em>type</em> of problem but almost certainly not the same <em>magnitude</em>. The German-language WikiProjekt on AI and Wikipedia itself notes that it remains \"unsatisfactorily answered\" (<em>unbefriedigend beantwortet</em>) whether generative AI is actually undermining Wikipedia's quality-control mechanisms, or whether LLM-generated articles are being caught reasonably well alongside the existing, never-ending stream of human-produced vandalism, hoaxes, and poorly sourced content <sup class=\"trigger\"><a href=\"https://de.wikipedia.org/wiki/Wikipedia:WikiProjekt_KI_und_Wikipedia/Wikiversum#:~:text=Unbefriedigend%20beantwortet%20bleibt%2C,gut%20erkannt%20werden.\" target=\"_blank\">1</a><span class=\"tooltip top center citation\"><a href=\"https://de.wikipedia.org/wiki/Wikipedia:WikiProjekt_KI_und_Wikipedia/Wikiversum#:~:text=Unbefriedigend%20beantwortet%20bleibt%2C,gut%20erkannt%20werden.\"><span class=\"source\"><img alt=\"\" src=\"https://p.kagi.com/proxy/favicons?c=lQ30-cxV_E2Vdyqb6hrbuYgId7qu7hecjO7o1T7aqQDFc7XpcLttL1Ryala-GNHJAQ-fjt8WMTZhLN6U1wzeZKYM0S0jk4QmuuY129Xs3iY%3D\"> de.wikipedia.org</span><b>1. Wikipedia:WikiProjekt KI und Wikipedia/Wikiversum – Wikipedia</b></a><span class=\"passage\">Unbefriedigend beantwortet bleibt, ob die generative KI die Qualitätskontrollmechanismen von Wikipedia unterläuft - oder ob diese LLM-generierte Artikel neben dem bestehenden nie endenden Strom von vom Menschen erzeugtem Vandalismus, Hoaxes oder Artikeln mit fehlenden oder irreführenden Referenzen gut erkannt werden.</span></span></sup>.</p>\n<h3>Why the scale differs</h3>\n<p>Several structural factors make English Wikipedia the primary target:</p>\n<table>\n<thead>\n<tr>\n<th>Factor</th>\n<th>English Wikipedia</th>\n<th>German Wikipedia</th>\n</tr>\n</thead>\n<tbody>\n<tr>\n<td><strong>Article count</strong></td>\n<td>~6.9 million</td>\n<td>~2.9 million</td>\n</tr>\n<tr>\n<td><strong>New articles/day</strong></td>\n<td>Substantially higher volume</td>\n<td>Lower volume</td>\n</tr>\n<tr>\n<td><strong>Language accessibility</strong></td>\n<td>English is the default output language for most LLMs; global lingua franca</td>\n<td>German is competently handled by LLMs, but fewer non-German-speakers attempt to contribute</td>\n</tr>\n<tr>\n<td><strong>Global South contributors</strong></td>\n<td>Large and growing; LLM use sometimes driven by good intentions but poor execution</td>\n<td>Much smaller pool of non-native contributors</td>\n</tr>\n<tr>\n<td><strong>New Page Patrol backlog</strong></td>\n<td>Chronic, often months-long backlog — a well-known bottleneck</td>\n<td>Tighter review pipeline; new articles are screened more quickly</td>\n</tr>\n<tr>\n<td><strong>Paid/promotional editing</strong></td>\n<td>Enormous and persistent problem, now increasingly AI-assisted</td>\n<td>Present but smaller in scale</td>\n</tr>\n</tbody>\n</table>\n<p>The last point about new page review is particularly important. English Wikipedia's NPP backlog has been a systemic issue for years, and the flood of AI-generated articles exacerbates an already strained system. German Wikipedia's review infrastructure, while not immune, handles a smaller incoming volume and has historically kept closer to real-time review.</p>\n<h3>What the German community <em>does</em> worry about</h3>\n<p>Even though the raw volume problem is smaller, the German debate revealed concerns that overlap with English Wikipedia's experience:</p>\n<ul>\n<li><strong>AI slop</strong> (<em>KI-Müll</em>) entering the encyclopaedia undetected, particularly where it mimics plausible but unsourced or subtly wrong content <sup class=\"trigger\"><a href=\"https://de.wikipedia.org/wiki/Wikipedia:WikiProjekt_KI_und_Wikipedia/Regeln#:~:text=Ziel%20der%20Sensibilisierung%2C,mehreren%20Textteilen.%20Humans\" target=\"_blank\">2</a><span class=\"tooltip top center citation\"><a href=\"https://de.wikipedia.org/wiki/Wikipedia:WikiProjekt_KI_und_Wikipedia/Regeln#:~:text=Ziel%20der%20Sensibilisierung%2C,mehreren%20Textteilen.%20Humans\"><span class=\"source\"><img alt=\"\" src=\"https://p.kagi.com/proxy/favicons?c=lQ30-cxV_E2Vdyqb6hrbuYgId7qu7hecjO7o1T7aqQDFc7XpcLttL1Ryala-GNHJAQ-fjt8WMTZhLN6U1wzeZKYM0S0jk4QmuuY129Xs3iY%3D\"> de.wikipedia.org</span><b>2. Wikipedia:WikiProjekt KI und Wikipedia/Regeln – Wikipedia</b></a><span class=\"passage\">Ziel der Sensibilisierung, wenn KI als Hilfsmittel verwendet wird, ist der Hinweis auf Fehlermöglichkeiten und die Verhinderung, das AI Slop (KI Müll) Eingang in die Wikipedia findet. Begriffe: Ein Textteil sind zusammengehörige inhaltliche Aussagen, die belegt (REF) sein müssen und eine Quelle, ein Werk, richtig darstellen. Ein Artikel besteht in der Regel aus mehreren Textteilen. Humans ...</span></span></sup>.</li>\n<li><strong>Existing reviewer fatigue</strong> being compounded — even a modest increase in low-quality submissions matters when the reviewer pool is already stretched.</li>\n<li><strong>Detection difficulty</strong> — the same fundamental problem that English Wikipedia faces. German Wikipedia's smaller scale doesn't make individual AI-generated edits easier to spot.</li>\n</ul>\n<h3>Takeaway</h3>\n<p>The German Meinungsbild was arguably passed <em>preventively</em> — before the problem reached English Wikipedia–like proportions — rather than reactively. The community decided it was better to establish a clear norm early than to wait until the volume became unmanageable. That pre-emptive logic was explicitly cited by proponents during the debate.</p>\n<p>English Wikipedia, by contrast, is already deep into the reactive phase, where the volume of AI-generated submissions has materially outpaced the community's review capacity — particularly at NPP and in topic areas with few active subject-matter experts.</p>","md":"<details><summary>Thinking</summary>\n\nThe user is asking whether German Wikipedia faces the same scale problem as English Wikipedia with AI-generated content overwhelming reviewer capacity. Let me work with what I know and what the research has found.\n\nKey contextual differences between the two projects:\n\n1. **Scale**: English Wikipedia is far larger (~6.9 million articles vs ~2.9 million on German Wikipedia) and receives far more new article submissions, including from the Global South where LLM-generated content in English is a particular vector.\n\n2. **New page patrol**: English Wikipedia has a massive backlog problem with new page patrol (NPP), which is where most AI-generated articles first appear. German Wikipedia has a different new article review process.\n\n\n3. **Language barrier**: English is the lingua franca and the primary output language for most LLMs, making English Wikipedia a more attractive target. German Wikipedia benefits from some insulation through language, though LLMs are increasingly competent at generating German text.\n\n4. **The Meinungsbild itself**: The near-blanket ban that German Wikipedia passed indicates the community saw a problem worth addressing, yet the research suggests whether AI content is actually overwhelming quality control remains an open question.\n\n</details>\n\n## AI content volume: German vs English Wikipedia\n\n### The short answer\n\nGerman Wikipedia faces the same *type* of problem but almost certainly not the same *magnitude*. The German-language WikiProjekt on AI and Wikipedia itself notes that it remains \"unsatisfactorily answered\" (*unbefriedigend beantwortet*) whether generative AI is actually undermining Wikipedia's quality-control mechanisms, or whether LLM-generated articles are being caught reasonably well alongside the existing, never-ending stream of human-produced vandalism, hoaxes, and poorly sourced content [^1].\n\n### Why the scale differs\n\nSeveral structural factors make English Wikipedia the primary target:\n\n| Factor | English Wikipedia | German Wikipedia |\n|---|---|---|\n| **Article count** | ~6.9 million | ~2.9 million |\n| **New articles/day** | Substantially higher volume | Lower volume |\n| **Language accessibility** | English is the default output language for most LLMs; global lingua franca | German is competently handled by LLMs, but fewer non-German-speakers attempt to contribute |\n| **Global South contributors** | Large and growing; LLM use sometimes driven by good intentions but poor execution | Much smaller pool of non-native contributors |\n| **New Page Patrol backlog** | Chronic, often months-long backlog — a well-known bottleneck | Tighter review pipeline; new articles are screened more quickly |\n| **Paid/promotional editing** | Enormous and persistent problem, now increasingly AI-assisted | Present but smaller in scale |\n\nThe last point about new page review is particularly important. English Wikipedia's NPP backlog has been a systemic issue for years, and the flood of AI-generated articles exacerbates an already strained system. German Wikipedia's review infrastructure, while not immune, handles a smaller incoming volume and has historically kept closer to real-time review.\n\n### What the German community *does* worry about\n\nEven though the raw volume problem is smaller, the German debate revealed concerns that overlap with English Wikipedia's experience:\n\n- **AI slop** (*KI-Müll*) entering the encyclopaedia undetected, particularly where it mimics plausible but unsourced or subtly wrong content [^2].\n- **Existing reviewer fatigue** being compounded — even a modest increase in low-quality submissions matters when the reviewer pool is already stretched.\n- **Detection difficulty** — the same fundamental problem that English Wikipedia faces. German Wikipedia's smaller scale doesn't make individual AI-generated edits easier to spot.\n\n### Takeaway\n\nThe German Meinungsbild was arguably passed *preventively* — before the problem reached English Wikipedia–like proportions — rather than reactively. The community decided it was better to establish a clear norm early than to wait until the volume became unmanageable. That pre-emptive logic was explicitly cited by proponents during the debate.\n\nEnglish Wikipedia, by contrast, is already deep into the reactive phase, where the volume of AI-generated submissions has materially outpaced the community's review capacity — particularly at NPP and in topic areas with few active subject-matter experts.","profile":{"id":"1b6d99f4-1649-48c8-8ee3-d6774318f814","name":"wikimedia-policy","accessible":true,"model":"claude-4-6-opus-thinking","deprecate":false,"retired":false,"successor_model":null,"model_name":"Claude 4.6 Opus (reasoning)","model_provider":"anthropic","model_input_limit":1000000,"recommended":true,"model_info":"\n\n\n\n<div class=\"heading\">\n <div class=\"left\">\n \n \n \n <svg class=\"model-icon icon-xs\" data-model=\"claude-4-6-opus-thinking\">\n <use href=\"#icon-sparkles\"/>\n </svg>\n <h3>\n \n <span class=\"model-name\">wikimedia-policy</span>\n </h3>\n </div>\n\n \n <a href=\"/settings?p=custom_assistant&id=1b6d99f4-1649-48c8-8ee3-d6774318f814\">\n <svg class=\"icon-sm edit-icon\" aria-hidden=\"true\">\n <use href=\"#icon-pencil\" />\n </svg>\n </a>\n \n</div>\n\n\n<hr />\n<ul>\n \n <li>\n <h4>\n Model\n </h4>\n <span class=\"flex align-center gap-8\">\n <svg class=\"icon-xs icon\">\n <use href=\"#icon-anthropic\" />\n </svg>\n <span class=\"model-name\">Claude 4.6 Opus (reasoning)</span>\n </span>\n </li>\n <li>\n <h4>\n Web access\n </h4>\n <span>On</span>\n </li>\n <li>\n <h4>\n Lenses\n </h4>\n <span></span>\n </li>\n \n</ul>\n\n\n","model_provider_name":"Anthropic","internet_access":true,"personalizations":false,"lens":{"id":21245,"name":null},"shortcut":null,"is_default_profile":false},"metadata":"<li>\n <span class=\"attribute\">Model</span>\n <span class=\"value\"><i class=\"icon-xs mr-4\"><svg><use href=\"#icon-anthropic\"></use></svg></i>Claude 4.6 Opus (reasoning)</span>\n</li>\n<li>\n <span class=\"attribute\">Version</span>\n <span class=\"value\">claude-opus-4-6</span>\n</li>\n<li>\n <span class=\"attribute\">Speed (tok/s)</span>\n <span class=\"value\">36</span>\n</li>\n<li>\n <span class=\"attribute\">Tokens</span>\n <span class=\"value\">4,552</span>\n</li>\n<li>\n <span class=\"attribute\">Cost / Total ($)</span>\n <span class=\"value\">0.06 / 0.13</span>\n</li>\n<li>\n <span class=\"attribute\">End to end time (s)</span>\n <span class=\"value\">35.84</span>\n</li>\n<li>\n <span class=\"attribute\">Submitted</span>\n <span class=\"value\">2026-02-19 18:11</span>\n</li>\n\n","documents":[]}]