2025/02/01 by Eric Deibel
Economics, Econometrics and Finance · Social Sciences · #Economic and Technological Innovation #Ethics and Social Impacts of AI #Information Systems Theories and Implementation
paper · doi:10.1177/02704676251353108
openalex publication_date 2025/02/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/29
The article examines the superimposition of the language of political theory on AI. This refers to values that are moralistic to the extend that they suggest what AI ought to be, more open, transparent, secure, fair, trustworthy and so on. Firstly, theoretical observations are made about the STS of the 80 s and early 90s: on Winner and Latour as social contract theorists. By extension STS can be seen as a relatively consistent approach to social contract theorizing, whether as Winner's technical constitution, Latour's nonmodern constitution, or otherwise. Secondly, the approach is applied to AI, which is to say that a variety of cases will be discussed that are selected for how they invoke the social contract, in terms of the conventional concepts like legitimacy, security, property, truth, freedom and so on. Included are ChatGPT, OpenAI, Sovereign AI, Constitutional AI, among others. The aim of the paper is, therefore, methodological; establish proximity to cases but remain detached from the hype that surrounds AI. What this requires is a broader perspective on the relation of STS to political theory (part 1) and its application (part 2) to a variety of claims on what makes AI responsible (as in responsible AI) rather than disruptive, open (as in Open AI) rather than opaque, “human-centric” rather than being excessively about nonhumans.