Contemporary debates about artificial intelligence (AI) still treat automation as a straightforward substitution of human labor by machines. Drawing on Goffman’s dramaturgical sociology, this paper reframes AI in the workplace as supplementary rather than substitutive automation. We argue that the central—but routinely overlooked—terrain of struggle is symbolic-interactional: workers continuously stage, conceal, and re-negotiate what counts as “real” work and professional competence. Large language models (LLMs) such as ChatGPT exemplify this dynamic. They quietly take over the invisible, routinised tasks that underpin cognitive occupations (editing, summarizing, first-draft production) while leaving humans to enact the highly visible or relational facets that sustain occupational prestige. Drawing on diverse sources to illustrate our theoretical argument, we show how individual workers, dramaturgical teams, and entire professional fields manage impressions of expertise in order to counter status threats, renegotiate fees, or obscure the extent of AI assistance. The paper itself, having been intentionally written with the ‘aid’ of all presently available frontier AI models, serves as a meta-reflexive performance of professional self-staging. The dramaturgical framework clarifies why utopian tales of friction-free augmentation and dystopian narratives of total displacement both misread how automation is actually unfolding. By foregrounding visibility, obfuscation, and impression management, the article presents a differentiated case for AI’s impact on the performative structure of work, outlines diagnostic tools for assessing real-world AI exposure beyond hype-driven headlines, and argues for a more human-centered basis for evaluating policy responses to the ‘fourth industrial revolution.’ In short, AI enters the labor process not as an autonomous actor, but as a prop within an ongoing social performance—one whose scripts, stages, and audiences remain irreducibly human.
Klowait et al. (Thu,) studied this question.