"Don't Blink": Protecting the Wikimedia model, its people, and its values in September 2026. 29 October 2026 by Allison Davenport, Lead Public Policy Analyst, Wikimedia Foundation Welcome to "Don't Blink"! Every month Wikimedia Foundation, Inc. share public policy developments from around the world that shape people's ability to participate in the free knowledge movement. In case you blinked last month, here are the most important advocacy topics that have kept the Wikimedia Foundation busy. The Global Advocacy team works to advocate laws and government policies that protect the volunteer community-led Wikimedia model, Wikimedia's people, and the Wikimedia movement's core values. To learn more about Wikimedia Foundation, Inc. and the work Wikimedia Foundation, Inc. do with the rest of the Foundation: follow Wikimedia Foundation, Inc. on LinkedIn, X (formerly Twitter), and Bluesky; sign up for its quarterly newsletter or Wikimedia public policy mailing list; and visit its Meta-Wiki webpage. Representing human knowledge at the United Nations General Assembly (UNGA) High-Level Week 2026 [Read its Foundation statement about participating at UNGA]. Every year in September, the United Nations General Assembly is convened in New York City to debate and make decisions about questions of global importance. During the 81st session of UNGA in September 2026, the topic of artificial intelligence (AI) and its impact on the world were at the forefront of conversations. While many of these discussions focused on recent, eye-catching headlines about AI safety, the Wikimedia Foundation brought to the table its own experience about content reuse and the challenges that AI is posing for open knowledge infrastructure. Bernadette Meehan (CEO of the Foundation) joined several talks throughout the week to emphasize three key messages about protecting free knowledge online: * Human knowledge is essential: Knowledge created and curated by humans serves as the cumulative memory of all humanity, and the standards around verifiability, neutrality, and transparency that organize the Wikimedia projects serve to underpin the success and accuracy of AI models. * Wikipedia is free, but infrastructure is not: While the content on Wikipedia is free and open for anyone to use and build upon with the proper attribution, the cost of rising web scraping and automated bot traffic on the infrastructure hosting the projects must be addressed. It is crucial in order to secure the long-term sustainability of the online information ecosystem. * AI safety is important, but the AI governance and regulation conversation need to be broader: Lawmakers reasonably want to address the potential safety risks of AI systems, but this alone is not enough to protect the world from the rest of AI's negative effects. They also need to address how to protect human knowledge sources from AI, and include stakeholders from civil society, multilateral institutions, academia, and news media in discussion rather than only a handful of high-profile corporations. The Foundation will continue showing up at these important conversations to make sure that the effects of AI on open knowledge are properly understood and taken into account by those shaping the future of AI governance. Explaining how online age verification affects the Wikimedia projects at the Future of Speech Online Conference [Watch its panel participation on age verification and its impact on anonymity]. The 2026 Future of Speech Online Conference, hosted by the Center for Democracy and Technology (CDT), examined the state of anonymity online and how current technological, political, and regulatory changes may threaten protections for anonymity in the future. A series of panels with experts from civil society, academia, and technology companies explored the importance of anonymous participation to freedom of expression and privacy, how AI tools and the reaction to them risk eroding anonymity online, and ways to maximize protections for anonymity while preventing harm online. Ricky Gaines (Human Rights Policy and Advocacy Lead) joined a panel about online age verification to share the Foundation's concerns about the trend of age gating requirements being included in laws related to child safety. Ricky discussed how collecting additional information to meet these requirements could negatively affect participation on the Wikimedia projects. The Wikimedia model has historically preserved anonymous and pseudonymous - as in under a user name - participation on the projects, since this plays a vital role in allowing people around the world to contribute to and read about topics on Wikipedia that may be sensitive and/or inspire political or interpersonal harassment and retaliation. In addition to the implications that age verification has for free expression, Ricky emphasized that these changes would also present an unreasonable compliance burden for the Foundation - since it currently collects minimal personally identifiable information (PII) about volunteer contributors and readers, and would be required by the laws to collect information about both child and adult users. All of the panels from the conference can be viewed on CDT's event webpage. Contributing with Wikimedia Australia (WMAU) to Australian privacy reform consultation [Read its joint response to the consultation and Wikimedia Australia's blog post about it]. Wikimedia Australia and the Foundation provided a joint response to a consultation regarding the Australian government's proposed Privacy Amendment (Personal Data Protection) Bill 2026, a comprehensive update to Australia's privacy laws relating to digital technology. Wikimedia Foundation, Inc. support a privacy framework that gives people meaningful control over their personal data, promotes data minimization, and addresses harmful data practices like the excessive collection and/or monetization of personal data. However, Wikimedia Foundation, Inc. also have concerns that certain parts of the draft law are so broad that they could have unintended consequences for the open knowledge ecosystem and other public interest projects online. In particular, its consultation response provided several recommendations about the proposed "right to erasure" in the law, which allows individuals to request that all data about them be deleted. Its submission highlighted how the maintenance of transparent records - such as revision histories, talk pages, and other artifacts of collaborative editing - are necessary for the continued operation of free and open knowledge projects like Wikipedia. While the law recognizes some important exceptions from this duty for platforms, Wikimedia Foundation, Inc. advocated additional exceptions that would protect the integrity of publicly available knowledge resources and the sources on which they rely from bad faith attempts to suppress accurate information. Wikimedia Foundation, Inc. believe that it is possible to pass privacy reform that appropriately balances personal protections with preserving the public interest infrastructure that gives Australians access to trustworthy, free and open knowledge. Championing free knowledge at WikiConference India (WCI) 2026 and beyond [Read more about WCI 2026]. India is a central focus of the Wikimedia Foundation's global public policy and advocacy efforts given its scale, regulatory dynamism, and the vibrancy of its Wikimedia volunteer communities. As one of the largest user bases for Wikimedia projects, developments in India's policy landscape have far-reaching implications not only for local contributors, but also for global norms on intermediary liability protections, platform governance, and access to knowledge. Recognizing the importance of bringing community perspectives to policy conversations in the country, Rachel Arinii Judhistari (Lead Public Policy Specialist for Asia) joined WikiConference India (WCI) in September 2026 to present a session about "Navigating India's Policy Landscape." Rachel outlined public policy priorities for free knowledge in India, shared key messages that Wikimedians can bring to lawmakers, and discussed the Global Advocacy team's current and upcoming engagements in the country. Session attendees learned how to build relationships and partnerships in order to become champions of Wikimedia's volunteer-led, public interest model. Rachel, along with other staff from the Foundation, have also been conducting outreach to lawmakers and ministers, sharing how the Wikimedia projects can contribute to government goals related to education and the preservation of local languages and culture. They recently met with several representatives of the government in Keralam (previously Kerala), including Member of Parliament Hibi Eden, Member of Legislative Assembly Chandy Oommen, former Director of General Education Snehil Kumar Singh, and Chief Minister V. D. Satheesan. These meetings helped introduce the government to Wikimedia projects and the broader open knowledge ecosystem, providing opportunities to collaborate on projects related to educational resources in the Malayalam language and also school initiatives around digital literacy. Meetings and collaborations like these help to shed light on how the Wikimedia projects differ from platforms like social media, and why they should be protected when making policy decisions about regulating the internet. Can you help Wikimedia Foundation, Inc. translate this article? In order for this article to reach as many people as possible Wikimedia Foundation, Inc. would like your help. Can you translate this article to get the message out?

Wikipedia bars LLM-written articles since March 2026 as AI cleanup continues. ColdCaseTech Author Originally published on ColdCaseTech. Disclosure: this post contains affiliate links. ScaleUp Web may earn... Disclosure: this post contains affiliate links. ScaleUp Web may earn a commission at no extra cost to you. Then vs. Now (News report): Then: WikiProject AI Cleanup states that the Wikipedia guideline WP:LLM has prohibited using LLMs to generate or rewrite article content since March 2026, with limited exceptions for translation and basic copyediting. Now: Founding member Ellia Lebleu told IEEE Spectrum that Pangram flagged perhaps 20 percent of new articles and drafts as AI-generated in late 2025 and 15 or 16 percent in early 2026, and that it is too early to tell whether the rule is working. WikiProject AI Cleanup, the volunteer effort to find and remove AI-generated material from Wikipedia, states that the guideline WP:LLM has prohibited the use of large language models to generate or rewrite article content since March 2026, with limited exceptions for translation and basic copyediting (Wikipedia). In an interview published on 2026-10-21, founding member Ellia Lebleu told IEEE Spectrum that the detection company Pangram flagged perhaps 20 percent of new articles and drafts as AI-generated in late 2025, and 15 or 16 percent in early 2026 (IEEE Spectrum). Key facts. | Item | Detail | Source | | Policy | WP:LLM prohibits LLMs to generate or rewrite article content since March 2026 | Wikipedia | | Exceptions | Translation and basic copyediting | Wikipedia | | Late 2025 estimate | Pangram flagged perhaps 20 percent of new articles and drafts | IEEE Spectrum | | Early 2026 estimate | Pangram flagged 15 or 16 percent | IEEE Spectrum | | Tool in development | Wikimedia Foundation is working on Paste Check | IEEE Spectrum | Timeline. * 2022-11-30: ChatGPT was released, the date before which article text is very unlikely to be AI-generated (Wikipedia). * 2023-12-04: The first founding members added their names to the project participant list (Wikipedia). * 2026-10-21: IEEE Spectrum published its interview with Lebleu (IEEE Spectrum). Details (as reported). The project's stated goals include identifying AI-written text, identifying AI-generated images, keeping track of AI-using editors, educating editors, and protecting against disruption from AI agents (Wikipedia). A lot of LLM-generated text inserted from 2022 to 2026 remains on the site, according to the project page, and pages clearly entirely LLM-generated can be nominated for speedy deletion under WP:G15 (Wikipedia). Early articles repeated phrases such as "stands as a testament" or "rich cultural heritage", while by 2025 the signs had become subtler and more structural (IEEE Spectrum). Lebleu called fake or misleading citations a major worry, adding that AI more often finds a real source but gets its content completely wrong (IEEE Spectrum). Lebleu explained that Pangram was tested on Wikipedia content from before ChatGPT and found no false positives, so its results provide a very conservative lower bound (IEEE Spectrum). The vast majority of AI-generated content is submitted by beginner editors, Lebleu added, and the rule has been made visible in welcome messages, editing help and advice (IEEE Spectrum). On tooling, another Wikipedia editor has built an AI source verifier that checks whether the sources on a page match the claims, and the Wikimedia Foundation is developing Paste Check to flag text pasted directly from some AI chatbots (IEEE Spectrum). The Foundation is also working on an attribution framework that could show AI users when an article was last edited, how comprehensive it is and whether editors have flagged problems (IEEE Spectrum). Keep one backup somewhere else. A copy of critical data with a second provider means one compromised account or one regional outage cannot take everything down with it. See Contabo storage plans (affiliate link) What is not yet confirmed. * Lebleu stated that it is still a little early to tell whether the rule prohibiting AI writing is working (IEEE Spectrum). * It is not yet known whether the drop in flagged content reflects the recent policies or models getting more powerful before Pangram has had time to adapt, according to Lebleu (IEEE Spectrum). * Because the AI source verifier is itself an AI model, there is no guarantee it will be correct every time, Lebleu stated (IEEE Spectrum). Sources. Tools related to this case. * A standby server outside your main cloud for failover: Contabo dedicated servers * Routing optimizer for unstable or high-latency connections: GearUP * Keeps routers and home servers up through short power cuts: UPS battery backups These are affiliate links: ScaleUp Web may earn a commission at no extra cost to you. As an Amazon Associate, ColdCaseTech earns from qualifying purchases. ColdCaseTech is an independent record of technology's promises: every line links to its source. New cases daily at coldcasetech.com. Follow on X, Bluesky or Telegram. ColdCaseTech Technology contributor sharing insights via Dev.to.