Academics Have a Duty to Defend Wikipedia From Generative AI, Argues LSE Impact Blog Post
Academics have an ethical duty to defend open knowledge systems such as Wikipedia from generative AI companies that divert its readers and misappropriate volunteer labor, argues an essay published on the LSE Impact Blog on 19 October 2026. The post, by Helen Choi, contends that because academia participates in and benefits from knowledge production, it bears responsibility for the structures that sustain Wikipedia — and that higher education institutions must change how they evaluate faculty if that duty is to be met.
The essay draws on Choi's article, "Towards a practice of knowledge commoning: academics' ethical duties to defend Wikipedia in the age of generative AI", published in AI and Ethics. The piece was submitted to Hacker News by user jruohonen and drew 158 points and 93 comments.
The core argument: AI products strip Wikipedia of the readers who become its editors
Wikipedia's existential risk comes from a readership-to-editor pipeline that generative AI is breaking. When people use AI products such as Google's Gemini or ChatGPT, they receive immediate answers. Those answers sometimes include links to Wikipedia, but most users do not click them, choosing instead to accept machine-generated output without verification. Choi cites research indicating that if this behavior continues, Wikipedia readership will continue to decline.
The consequence is structural, not merely financial. Wikipedia is run by a volunteer community that writes, codes, and administers the site, and that community is drawn from Wikipedia's readers. Fewer readers means fewer articles, updates, and corrections — and ultimately, Choi writes, a "derelict and quite possibly defunct encyclopedia." She argues this would be a loss for humanity and for the AI industry itself, which relies heavily on Wikipedia and its five billion words to train its systems.
Choi frames Wikipedia as more than training data: it is a self-governing community of people that exists independently of commercial or governmental controls, producing and sharing knowledge for the common good. She notes it has outperformed commercial efforts such as Microsoft's Encarta and xAI's machine-generated Grokipedia, and that it is perceived as influential enough to have been banned in countries including China.
The complaint against AI companies: scraping, licensing, and profit from volunteer labor
The essay distinguishes between two ways AI companies take Wikipedia content, and treats both as problematic for a weakened volunteer community. Some companies take Wikipedia's content through unauthorised scraping. Others — including Microsoft, Amazon, and Meta — pay Wikipedia for high-speed API access to its pages.
Choi's characterization is that these companies have "misappropriated the hard work of volunteers for their own profit, reducing writing and research to machine feed." The essay does not treat paid licensing as a resolution of the underlying problem; it presents both scraping and licensing as part of the same dynamic in which AI companies control online search and pose existential challenges to a "weakened and demoralised" volunteer community.
Why academia specifically owes a duty: Iris Marion Young's Social Connection Model
Academia's obligation is grounded in its position within knowledge production, not in any claim that academics caused the harm. Choi invokes Iris Marion Young's Social Connection Model, under which a party's duty to remedy injustice is determined by its connection to the processes and structures in which it participates. As a participant in knowledge processes, academia has a responsibility to address injustices in those processes — including those committed by the AI industry against vital sources of knowledge like Wikipedia.
Choi adds a capacity argument: academia has access to resources and higher status than the largely anonymous group of Wikipedia volunteers, giving it both the power and the capacity to act. As people trained in producing knowledge and who teach the next generation of knowledge producers, academics are, in her framing, uniquely qualified to support Wikipedia's current and future reserves of knowledge and its volunteer community. She is explicit that academics are not the ones who harmed Wikipedia — they are fellow stewards of knowledge who have a duty to help defend it from those who did.
How academics can support Wikipedia, according to the essay
Choi proposes four concrete channels through which academics can act, each tied to skills they already have.
Editing directly. Subject-matter expertise can be used to address content gaps, correct errors, and provide context on complicated topics. Access to and facility with high-quality sources can improve Wikipedia's quality. Because Wikipedia's citation practices mirror academic practices, academics may find editing Wikipedia less daunting than other newcomers do.
Citing Wikipedia in scholarly work. Choi argues that just as academics must cite any AI-generated machine outputs they rely on, they can cite Wikipedia when they use it. She says this highlights Wikipedia's value to other academics and increases transparency in research processes. She notes that despite longstanding skepticism among some scholars, citations to Wikipedia in academic journals appear to be increasing, including in high-impact health sciences publications such as The BMJ.
Teaching students to contribute. This builds knowledge about Wikipedia in future generations while strengthening students' subject-matter knowledge and information literacy. Choi cites the non-profit Wiki Education Foundation, through which more than 152,000 students at over 8,000 colleges and universities in the United States and Canada have contributed more than 125 million words to Wikipedia since 2010. She reports that students find writing for Wikipedia motivating because they write for a real-world audience, and that they develop practical information literacy skills in source evaluation, copyright compliance, and source attribution. Because Wikipedia currently bans AI-generated writing, Choi suggests assigning Wikipedia work is one way for students to build writing and thinking skills independently. She notes that about 19% of Wikipedia's new active editors come from Wiki Education's programs.
Institutional recognition. Choi argues the ethical obligation cannot be fulfilled by individual academics alone. Most academics are evaluated on scholarly and research achievements, teaching performance, and service to their departments; public service such as contributing to Wikipedia goes largely unrecognized, if it is counted at all. She argues institutions must recognize Wikipedia contributions and provide resources and support. She links this to the public's eroding trust in higher education, suggesting a greater commitment to serving the public would reassert universities' role as stewards of knowledge that exist to serve humanity — a role she says is shared by Wikipedia.
The strongest counterargument: citing Wikipedia invites citogenesis
The most substantive objection in the Hacker News discussion targets the essay's recommendation that academics cite Wikipedia in their own work. A commenter who says they have contributed to Wikipedia and "seen how the sausage is made" called that advice "stupid," arguing that Wikipedia itself considers wikis — including Wikipedia — to lack sufficient authority to be cited as a source:
Such citations could be easily gamed (if you cite a particular change) or are useless (the content you cite may not be there in the future), and are obvious pathways to "citogenesis."
Citogenesis — the circular process by which a fact inserted into Wikipedia is cited by a publication and then re-cited back into Wikipedia as though the publication were an independent source — is a recurring theme in the thread. Another commenter framed the risk as accelerating an existing problem: "Citogenesis is already a considerable problem, can we not kick it into overdrive?"
The "turn tables" objection: academia once banned Wikipedia citations
Several commenters noted the reversal in academic attitudes. One wrote: "I'm old enough to remember when Wikipedia was not allowed to be cited in high school papers because it wasn't a sufficiently authoritative source. How the turn tables…" Another made the same point about higher education specifically: "Academia originally had issue with wikipedia due to the lack of academic oversight and therefore didn't allow it in citation for academic work. Now they have to defend it?"
A related comment invoked Wikipedia co-founder Larry Sanger: "I guess Larry Sanger was right all along about experts vs amateurs?"
The editor-retention objection: Wikipedia's problem predates AI
The essay attributes Wikipedia's editor shortage primarily to AI-driven readership decline. Multiple commenters argued the community's own culture is a larger and older cause. One long comment argued that Wikipedia's community often makes clear it does not want editors, and that people who tried to contribute "will tell you of the unjustifiable miserable time they had when wanting to do a good deed," calling this "a far larger, more evident, long-standing problem" than AI traffic diversion. The same commenter rejected the citation proposal outright:
Wikipedia being at risk is one thing, but asking for a cancer to be spread upon academia, I'm sorry but we are not at a point in which reputation can be squandered like that.
That commenter also argued the essay's own framing cuts against it, saying the acknowledgment that academia has lost public trust makes the proposal worse rather than better, because the reputational damage stems from the same political-bias and agenda-pushing criticisms frequently leveled at Wikipedia.
The "open to modify" objection: known authorship versus anonymous editing
A separate line of criticism questioned whether Wikipedia's editing model is defensible at all. One commenter argued that letting anybody edit any page makes it easy to insert knowledge of dubious quality and sources, and proposed an alternative:
We need "open access" knowledge systems with known authors who are sponsored by non-profit and educational organizations and its their duty to defend their reputation.
A related comment pointed to Wikipedia's sourcing policy rather than its openness, arguing it is hard to defend a system whose policies allow factually incorrect content to remain "just because it appears on a third-party source that the editors consider reliable."
Other threads in the discussion
- Scope of the title. One commenter objected to the phrase "open knowledge systems like Wikipedia," arguing that "like" excludes the named item, and that when the named item is the only important one, the distinction matters.
- Where the defense should be aimed. One commenter argued that defending open knowledge should mean attacking the monopoly of Elsevier and other publishers through new systems for peer-evaluated papers, "not by trying to resuscitate the body of the spook-filled Wikipedia."
- Alternative architectures. One commenter proposed that what academics need to defend is "wikipedia in latent space" rather than "walls of text that reflect the opinions of a few who have the time to edit," citing a project called latentpedia.org.
- Funding the commons. One commenter suggested AI companies should divert 5% of revenue to sustain Wikipedia and researchers.
- Faculty workload. One commenter who identified as a PhD-holding academic wrote simply: "We're overworked and don't have a duty to do anything."
- Access failures. One commenter noted that a CloudFront WAF blocked IPs from a Singapore ISP, calling it ironic for a post about defending open knowledge.
- Ideological bias claims. Two comments alleged ideological bias at Wikipedia, one stating that "Wikipedia is being lost to ideological bias" and another that "Wikipedia's bias against Israel is disgusting." Neither comment offered supporting evidence.
- A broader media critique. One commenter reflected on how news dissemination has changed, noting that an event in one small corner of the world can now reach anyone globally, and that commentary now arrives with bots mixed in.
- A general critique of academia. One long comment argued that academics should first explain "years of inaction" toward their students, and proposed cutting funding to programs where professors do not work in their field at least part-time. The comment did not engage with the essay's specific claims.
What the essay actually asks for
Stripped of the surrounding debate, Choi's proposal has two parts. The first is individual: academics should edit Wikipedia, cite it where they use it, and teach students to contribute. The second is institutional: universities should count public service such as Wikipedia contributions in faculty evaluation and back it with resources. The essay's central claim is that the duty follows from academia's connection to knowledge production rather than from any wrongdoing, and that Wikipedia's volunteer pipeline — which depends on readers becoming editors — is the specific mechanism that generative AI threatens.