Disruption of Economic Theories in the Era of Generative AI: Performance, Accountability, and Reward

Abstrakt

Generative artificial intelligence disrupts a foundational assumption underlying most economic theories of labor: that performance, accountability,
and reward are mutually intertwined. This idea paper problematizes that assumption and advances three working concepts for its analysis. The EMA framework (Empathy–Mechanics–Assistance) classifies labor activities according to the degree of human irreplaceability. The EMMA framework (Evaluation–Monitoring–Modification–Arbitration) describes what a responsible human worker concretely does in an environment where AI generates a substantial portion of outputs previously delivered by human effort alone. The Thesis on Responsible Cognitive Labor (TRCL) connects both frameworks into a working proposition: the value of human labor should be proportional to the degree of accountability assumed, not to the quantity of time expended or effort exerted. This proposition contains an empirically testable prediction: where accountability is institutionally anchored, human labor will retain its economic value longer than where such anchoring is absent. The paper acknowledges its limitations — the concepts are not yet sufficiently operationalized for formal testing and the predictions are directional rather than precise — and offers itself as a basis for further research rather than a finished theory.