Aerial vision is not simply a viewpoint. It is a machinic system in which humans, technologies, and institutions coordinate the production of visual authority from above. This article argues that aerial vision has always operated this way; as a distributing perceptual agency across sensing instruments, interpretive practices, and institutional structures, and that contemporary AI-driven systems intensify this condition by embedding perception within computational infrastructures that track, classify, and intervene without waiting for human judgment. Drawing on theories of aeriality, visuality, and media geography, integrated through Deleuze and Guattari's concept of the machinic assemblage, the article traces a genealogy across five phases: symbolic elevation, mechanical aeriality, institutional consolidation, the internalization of the aerial gaze within everyday spectatorship, and its automation through AI-driven sensing. These phases do not replace one another; they accumulate, each reorganizing what came before. What distinguishes the contemporary moment is not the scale of elevation but the increasing autonomy of these systems: AI-enabled drones and satellite networks no longer simply produce images for people to interpret; they feed data into systems that act. Indigenous and decolonial counter-aerial practices show that elevation can be reoriented toward relational and restorative witnessing. Understanding aerial vision as machinic reveals how contemporary visual infrastructures do not merely represent the world. They help govern it.
Lawrence Mullen (Fri,) studied this question.