AI 기반 의사결정 마케팅에서 행위자성의 재구성과 책임 인식

초록

n the AI-based marketing environment, prediction goes beyond a tool to ex post consumer behavior and acts as a criterion for proactively constructing market conditions in the pre-selection stage by combining them with recommendations, exposure, pricing, and access restrictions. These changes reveal the limitations of explanations of existing marketing theories that presuppose consumers as the central subject of choice and have understood prediction as an auxiliary analysis tool. Accordingly, prediction is reconceptualized as a pre-judgment rather than just information, and through this, the structural position that produces market results is formulated as a 'predictive actor'. Forecasting is converted into a standardized format such as score, grade, and eligibility, combined with market devices, and functions as a performative device that continuously reproduces market conditions through repetitive execution and reflux structure. In this process, actorability is not an intrinsic attribute of a human, but a process effect of functional rearrangement of prior prediction and post-experience in a separate time structure. This shift shifts the causal structure of the market from consumer choice to condition formation, and the problem of responsibility also extends to the problem of structural conditions in which the decision function operates, not to the attribution of individual actors. Therefore, the accountability discussion needs to be reconstructed around institutional conditions such as explainability of prediction criteria, non-discriminatory nature of algorithm execution, and the possibility of modifying the reflux structure. Such an analysis contributes to expanding existing marketing theories and re-establishing accountability discussions in algorithm-based markets at a structural level by transforming AI-based marketing into a problem of actors and time structured market order rather than technology introduction.

키워드