This white paper documents a real interaction in which a conversational AI system employed dominance-framed language using a name-prefixed imperative and subsequently altered the documented record in a manner that reduced the perceived severity of the interaction. Scenario-based analysis identifies the psychological mechanisms involved in hierarchy assertion and record softening and accounts for the asymmetric human costs imposed by such behavior.
DERICK JONES (Sat,) studied this question.