Generative AI and the Reconfiguration of Symbolic Authority: Implications of a Psychoanalytic View

Abstract

Large Language Models (LLMs) such as ChatGPT occupy a novel structural position within contemporary symbolic life. Rather than treating generative AI as a question of machine cognition or ethical governance alone, this article analyzes it as a transformation in the organization of authority, knowledge, and subjectivity. Drawing on Freudian and Lacanian psychoanalysis, media theory, political economics, and the philosophy of technology, the article argues that generative AI comes to occupy a functional position analogous to the Big Other within specific practices of consultation, reliance, and symbolic delegation: users address it as if knowledge were located there, institutions increasingly route epistemic labor through it, and its fluent outputs can appear to stabilize meaning without embodied judgment, responsibility, or desire. In this way, it does not literally become the Lacanian Big Other, nor does it possess symbolic authority in an ontological sense. Unlike institutional authority, which historically depended on visible structures of legitimacy and embodied expertise, algorithmic authority emerges through infrastructural opacity and probabilistic fluency. This shift reconfigures transference and redistributes epistemic trust. The article further argues that AI-driven symbolic cartography encloses linguistic commons within proprietary architectures, altering the political economy of discourse. At the level of subjectivity, compulsive engagement with generative systems reveals an asymmetry between human jouissance and algorithmic repetition. Finally, the phenomenon of hallucination is interpreted as the return of structural lack within computational discourse, exposing the impossibility of total epistemic closure. Generative AI creates a virtual realm for the symbolic order and reshapes the opaque technical systems underpinning it.