Robotoid humanness: when selfhood becomes machine-legible

Abstract As generative systems and socially responsive service agents enter everyday consumption, human–AI interaction increasingly resembles a quasi-social encounter rather than a utilitarian interface. This article theorizes how such encounters can reshape selfhood under conditions of continuous computational mediation. We introduce “robotoid humanness” to name an emergent drift in which consumers come to experience themselves as most fluent, correct, or socially viable when they become compatible with machine legibility, machine pacing, and machine logic. To explain how this drift can occur, we develop a three-stage mirroring mechanism: consumers enter a synthetic social reality that invites meaningful commitment; computational identity capture feeds back a reduced profile as personalized recognition, substituting a statistical abstraction for narrative self-understanding; and users adapt their self-presentation toward what the system can readily parse and reward, tightening alignment over repeated encounters. Grounding this account in the predictive-processing view of the self, we argue that what distinguishes AI-mediated mirroring from ordinary social looping is not the fact of feedback, but its character: where human interlocutors furnish heterogeneous and contestable evidence, algorithmic feedback is engineered to converge. We further show that the mechanism extends from predictive recommendation systems to open-ended, LLM-based conversational agents. The article repositions consumer-facing AI service agents as identity-relevant infrastructures, specifies testable propositions for empirical research, and articulates an autonomy risk that extends beyond privacy and bias: the normalization of reduced personhood as a standard of understanding in AI-mediated service life.