Abstract This paper argues that digital preservation institutions face a paradigm shift from entropic to adversarial threat models. Drawing on Weber’s method of ideal-typical construction, it distinguishes two orientations: the Archivist , who operates under what the paper terms institutional innocence — the historically warranted assumption that threats are passive, access is an unqualified good, and institutional continuity can be presumed — and the Fortress Commander , who designs for active, intelligent adversaries targeting cultural memory. The analysis traces this transformation across four fronts: cybersecurity, where ransomware campaigns exploit the trust architectures of open institutions; geopolitical instability, where preservation demands strategies of radical concealment and geographic distribution; biosecurity, where the pandemic revealed the invisible human infrastructure underlying all preservation as a complex operational achievement; and information warfare, where generative AI and coordinated disinformation erode the willingness of populations to distinguish truth from fabrication. Each front exposes a distinct logic through which the Archivist’s assumptions become vulnerabilities. The paper concludes with five operational imperatives and argues that preservation must now defend not only its contents but the epistemic conditions under which those contents retain their value.
Marcel Cotta (Fri,) studied this question.