Sustaining Community Archives through Student Research and Generative Artificial Intelligence

The presentation explores the evolution and impact of an open-access archive, Mapping Italian-Canadian Foodways — highlighting its origins, structure, and role within community-engaged digital pedagogy, along with recent efforts to extend this work through the integration of Generative Artificial Intelligence (GenAI). Situated at the intersection of sustainable digital humanities, global AI practices, and student-centred pedagogy, Foodways demonstrates how a community-based DH project can adopt emerging technologies while sustaining, over the long term, the pedagogical and cultural values that originally inspired it. Mapping Italian-Canadian Foodways (https://omeka.utm.utoronto.ca/s/foodways/page/home), hosted by the University of Toronto Mississauga, documents over one hundred years of Italian diaspora and culinary history across the province of Ontario, Canada. Emerging from community-engaged, student-driven research, the archive preserves the culinary heritage of Italian-Canadian communities through recorded oral histories, digitized artifacts, and historical- and geo-mapping. Designed as a sustainable pedagogical resource, the archive is continually expanded through student contributions in course-based assignments and research assistantships, and it serves as a reference guide in Italian cultural studies courses. Over five years of fieldwork have revealed a recurring challenge: minimal preservation of physical materials within the Italian-Canadian community, resulting in gaps in both documentation and representation. In response, Foodways is exploring how GenAI can act as a sustainable and collaborative extension of its archival and critical work. The project is currently testing GenAI tools to reconstruct incomplete historical narratives, generate descriptive metadata for under-documented materials, and visualize cultural patterns across time and space. These early experiments demonstrate how GenAI can support, rather than supplant, human interpretation and contribute to more dynamic, inclusive models of cultural preservation.