Operant Dyad: Defining the Task-Bound Working Relationship in Human-AI Systems

Abstract

This paper proposes the Operant Dyad as the configured, task-bound working relationship between one human and one functionally bounded adaptive AI system. The relationship forms through repeated correction: the human supplies goals, constraints, and defect reports; retained corrections shape later outputs; and the human inspects, accepts, rejects, or revises consequential results.

Preserved human authorship is the governing condition for a healthy Operant Dyad, not a feature presumed to exist in current systems. The paper defines Relationship State, distinguishes declared from inferred state, introduces Truxion as a proposed construct for recurring misalignment cost, and describes the Portable Operant Packet as a proposed continuity representation. It is a conceptual proposal and research agenda rather than an empirically validated system.