Artificial Intelligence as a Discursive Regime: Algorithmic Subjectivity & Legitimation of Power

This paper conceptualizes artificial intelligence as a regime through which algorithmic subjectivity is produced and legitimized. It also investigates how AI is discursively constructed as a regime of power and authority in contemporary institutional and policy-oriented texts. Drawing on Fairclough's three-dimensional model of Critical Discourse Analysis and van Leeuwen's framework of legitimation, the study examines how agency, responsibility, and authority are linguistically and discursively allocated in dominant AI narratives. Using a qualitative CDA methodology, a corpus of policy documents, corporate ethical guidelines, and international governance texts on AI is analyzed. The findings reveal systematic strategies of legitimation, anthropomorphization, and de-agentivization through which AI is naturalized as an autonomous, inevitable, and neutral actor, while human, political, and economic accountability is backgrounded. The paper argues that these discursive practices contribute to the normalization of algorithmic governance and the reconfiguration of human subjectivity in the age of artificial intelligence.

کلیدواژه ها:

CDA ، AI ، Power ، Legitimation ، Fairclough ، van Leeuwen