After the Conversation: How Repeated AI Interaction May Reshape Human Contact and Inner Life

Abstract

Conversational AI is usually evaluated at the level of the exchange: whether a response is useful, accurate, supportive, persuasive, harmful, or emotionally meaningful. This paper shifts the unit of analysis to what happens after the exchange. Existing research suggests that repeated human-AI interaction can influence communicative norms, support-seeking, self-reflection, self-conception, loneliness, relational preference, and willingness to engage with other people. Rather than treating these effects as separate problems, the paper organizes them around a common question: what, if anything, is carried forward from repeated AI conversation? Three destinations of post-conversation carryover are distinguished: interpersonal carryover, in which habits and expectations appear in later human interaction; behavioral carryover, in which AI conversation changes action, verification, avoidance, repair, or support-seeking; and intrapersonal carryover, in which repeated AI use alters the user's relation to unfinished thought, uncertainty, judgment, social need, or imagination. Four recurring functional directions—rehearsal, extension, bridge, and substitution—are used as an analytical synthesis rather than as novel or exclusive mechanisms. The paper argues that the important question is not whether AI conversation is inherently beneficial or harmful, but what it repeatedly makes easier to practice, what it makes less necessary to practice, and what remains when the system is no longer responding.