Envisioning Archival Images with Artificial Intelligence

Abstract

The literature review explores the role of Artificial Intelligence (AI) in enhancing access to and management of photographic archives. As digital and analog photographs proliferate in archival institutions, traditional approaches to organizing and describing these materials are increasingly inadequate. The review highlights the potential of AI, particularly computer vision (CV), to address the challenges associated with processing large volumes of digital images. CV algorithms, such as object detection and image classification, can automate tasks like image metadata generation, offering archivists new tools for organizing collections more efficiently. However, the adoption of AI in archival practice raises important ethical concerns, particularly regarding biases inherent in AI training datasets and technologies like facial recognition. Through various case studies, the review demonstrates that interdisciplinary collaboration between archivists, AI specialists, and scholars is crucial to developing effective AIdriven solutions. Projects like CAMPI and the Finnish Wartime Photograph Archive illustrate the practical benefits of AI, while emphasizing the need for archivists to develop AI and visual literacy. This review serves as a foundational resource for archival scholars and practitioners interested in utilizing AI to improve access to photographic archives.