Big AI’s demands for this world are becoming clearer. Microsoft, in 2023, announced plans to build new data centers with nuclear power to fuel new energy-hungry models (Calma, 2023). Google and Amazon made similar announcements subsequently (da Silva, 2024; Olick, 2024). Plans to build nuclear-powered AI data centers clearly illustrate the scale and consequences of AI as a social blueprint -- rendering clear “the choices (implicit or explicit) made in the course of technological innovation” and demanding reflection on “the grounds for making those choices wisely” (Winner, 1986, p. 18). Building on the material turn in Internet Studies (Hesmondhalgh, 2022; Sandvig, 2013), our preconference gathers scholars to explore ruptures against the growing cyberphysical project of “Big AI” (van der Vlist et al., 2024) or “AI as platform” (Mahnke & Bagger, 2024).
Our preconference, The Model and the Reactor, has three objectives:
1. Share findings and digital methods that expose AI’s global technological footprint with an emphasis on the Americas or engaged and speculative research on alternative AI infrastructures that may include local or regional infrastructure, the fediverse, frugal AI infrastructures, decentralized, and/or distributed infrastructures;
2. Facilitate comparative policy research on measures to promote alternative AI infrastructures as well as benefit public interest and community benefits for these alternative infrastructures;
3. Develop a joint statement about recommendations for a new infrastructure for AI to be written collaboratively by discussants.
Together, our pre-conference seeks to cultivate an international research community dedicated to understanding AI’s infrastructural impact and its alternatives. The conference offers international scholars a chance to develop collaborative projects as well as shape collective policy recommendations. Outputs directly advance the annual call for “strategies and tactics to address the ruptures caused by platformization” in this case of AI?
We focus our call on questioning what public interest infrastructure would look like for AI. Public interest AI refers to “support those outcomes best serving the long-term survival and well-being of a social collective construed as a ‘public’” (Public Interest AI, n.d.). The Paris Charter on Artificial Intelligence in the Public Interest (2025), published after the Paris AI Summit, aims to “encourage a more comprehensive and inclusive design of AI in the public interest, in terms of technology, organization and institutions that serve different jurisdictions and communities in attaining similar success.” Public interest AI, however, is already a contentious term and not dissimilar to other terms “AI for Good” or “Responsible AI” that can act as ethics washing (Bourne, 2024; Wagner, 2018). Scholarly attention is required to define public interest AI as a critical concept.
The preconference will be a full day to ensure there is time for presentations, networking, and collaborative activities. Participants will be selected through a peer-reviewed call with two tracks, one for presentations and a second for discussants and facilitators. Presentations will advance objectives 1 and 2 in the morning. The afternoon will leverage discussants and facilitators to develop collaborative research projects and synthesize research into a joint statement. Our schedule can accommodate two parallel tracks to welcome 20 presentations and a total of 50 participants. A closing plenary will showcase key themes, with discussants offering reflections and key insights from the day.
Our preconference is exceptionally multilingual, with collaborators working in English, French, Portuguese, and Spanish. The steering committee will accept abstracts in all four languages. We have also designed the pre-conference to accommodate multilingual submissions thanks to the support of the Chaire de recherche du Québec sur l’intelligence artificielle et le numérique francophones.
Organizers
Fenwick McKelvey is an Associate Professor in Information and Communication Technology Policy in the Department of Communication Studies at Concordia University.
Jonathan Roberge est professeur de l’Institut national de la recherche scientifique (INRS).
Mónica Humeres es Profesora Asistente del Departamento de Comunicación Social de la Universidad de Chile.
Claudia López works as an assistant professor at the Department of Informatics at Universidad Técnica Federico Santa María in Valparaíso, Chile.
Veridiana Domingos Cordeiro is a Professor at the Department of Sociology at the University of São Paulo (USP).
Luciano Frizzera is a Digital Humanities Postdoctoral Fellow at the University of Waterloo.
