Cognitive Pragmatics in Human–AI Interaction

Abstract

The rapid development of language-based artificial intelligence has transformed the ways humans interact, communicate, and construct meaning, particularly within social, educational, and personal reflective contexts. This transformation raises fundamental questions regarding how meaning is produced and interpreted when interaction no longer involves a human interlocutor. This study examines human–AI interaction through the lens of cognitive pragmatics, with a specific focus on users’ subjective experiences, inferential meaning-making processes, and relational negotiation. Adopting a qualitative phenomenological approach, the study explores participants’ lived experiences of interacting with language-based AI systems in academic and reflective settings. Data were collected through in-depth semi-structured interviews with active AI users and complemented by document analysis of selected interaction records. The data were analyzed using thematic analysis to identify recurring patterns of meaning construction and emergent pragmatic dynamics. The findings reveal three interrelated themes: the experience of an illusion of understanding, the attribution of intention and politeness to AI, and the negotiation of emotional distance accompanied by ethical reflection. These findings demonstrate that meaning in human–AI interaction is not derived from AI’s communicative capacities but is actively constructed through users’ pragmatic inference and metapragmatic awareness. Theoretically, this study extends the scope of cognitive pragmatics to the domain of human–technology interaction by foregrounding users’ experiential and inferential processes. Practically, the findings offer important implications for digital literacy, education, and the development of more reflective and responsible human–AI relationships