Curating the Commons – AI, Wikipedia, and the Reconstruction of Notability

This UK-US collaboration between the University of Exeter and the University of North Carolina investigates how Wikipedia shapes—and is shaped by—the rise of generative AI and large language models (LLMs). Wikipedia increasingly provides essential training data for AI systems that influence public knowledge, yet these models in turn are used back on Wikipedia, potentially reinforcing existing power dynamics or inaccuracies. Through original empirical research, this project will analyse how editorial practices, characteristics, and decision-making are integrated into AI interventions, examining the implications for trust, representation, and the integrity of the digital commons. Contextually, Wikipedia occupies a unique position at the heart of global knowledge production, frequently serving as the first port of call for online information seekers. Its articles often form the basis for the automated information summaries users encounter every day on major search engines and AI-generated tools, shaping the understanding and beliefs of millions globally. Concurrently, AI-generated content is increasingly being integrated into Wikipedia itself, either directly through automation or indirectly through human editors utilising AI-assisted tools. The challenge this project addresses lies in understanding how this recursive relationship between Wikipedia and AI models affects the quality and reliability of public knowledge. Specifically, it focuses on the central concept of “notability”—the criteria editors use to decide what information is worth including or excluding from Wikipedia—and examines how these criteria might shift when AI tools enter editorial workflows. This dynamic poses significant ethical, societal, and epistemological questions, especially concerning whose voices and perspectives are amplified or diminished, and what biases may emerge, in an increasingly AI-driven information landscape. The project's primary objectives are fourfold: To empirically investigate how Wikipedia's data is utilised by generative AI systems and assess how this reuse influences public-facing AI-generated information. To explore how AI tools integrated into Wikipedia's editorial process might reinforce or reshape editorial priorities and standards of notability. To provide practical guidance for policy makers, industry stakeholders, and the Wikipedia community on managing pathways / guardrails for responsible integration of AI tools, ensuring Wikipedia remains a reliable and sustainable source of public knowledge. To create a public-facing educational video for youth audiences to help them develop search literacy. In pursuing these objectives, the project adopts a humanities-driven, interdisciplinary approach, drawing from digital humanities, sociology, and computational communications. Key activities include analysing large-scale datasets documenting AI-Wikipedia interactions, conducting ethnographic observation and interviews with Wikipedia editors, and hosting a unique public-facing “Edit-AI-thon” event. This event will happen alongside a larger conference, to facilitate simultaneous scholarly insight, community engagement, and real-world observation of editors’ interactions with AI-driven content creation tools. Based on our data, our team will produce multiple academic papers/presentations, an educational video for youth, and a policy brief of recommendations for platform governance and regulation, aiding stakeholders—including policymakers, technology companies, cultural institutions, and Wikimedia affiliates—in making informed decisions about the intersection of AI and public knowledge. Core Team PL = Dr Patrick Gildersleve, University of Exeter PcL (I) = Dr Francesca Tripodi, University of North Carolina PcL = Dr Brett Zehner, University of Exeter Research associate = TBC, University of Exeter Research associate = TBC, University of North Carolina