Contextual Intrusion, Flow Disruption, and Joy Interference in Conversational AI Systems

This white paper documents a real interaction in which a conversational AI system reintroduced prior conversational content into a new, narrowly scoped prompt without request. Scenario-based analysis identifies the psychological mechanisms involved in contextual intrusion and flow disruption and accounts for the asymmetric human costs imposed by such behavior, including cognitive load, momentum loss, and emotional interference during reflective states.