AI-Mediated Miscommunication in a Multilingual, High-Context Society: Encoding, Decoding, and Meaning Distortion Among Bangladeshi Youth
Journal of Intelligent Digital Media and Communication · 2026
DOI: 10.67732/jidmc.v1i2.2 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Intelligent Digital Media and Communication, 2026. Verfasst von Md. Sipon Mia und 2 weiteren Autor:innen. Die Arbeit nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
This qualitative study examines how AI-generated messages produce miscommunication and meaning distortion among Bangladeshi youth in a multilingual, high-context communication environment. Most existing scholarship on AI-mediated communication is grounded in Western, English-dominant settings; how generative AI functions where multilingual code-switching and high-context cultural norms shape meaning remains largely unexamined. Drawing on Stuart Hall's Encoding-Decoding Model and Human-Machine Communication (HMC) theory, the study uses semi-structured in-depth interviews with 12 Bangladeshi youth and four expert informants, analysed through a theory-driven thematic analysis that also allowed data-driven subthemes to emerge inductively within the theoretical categories. Six interconnected themes emerged: instrumental, purpose-bounded AI use; identity, authenticity, and impression management; tone and emotional misalignment; linguistic hybridity and translation breakdown; cultural incongruence and Western-centric framing; and calibrated trust and critical reliance. Findings indicate that participants primarily use AI as an editing and refinement aid for their own personal, academic, and professional messages rather than as a communicative substitute, and routinely revise AI-generated text before sending it, positioning the human user as an active co-encoder rather than a passive recipient of AI output. Miscommunication occurred most often in tone, register, hierarchy, cultural framing, and perceived message provenance rather than lexical accuracy alone, indicating a recurring risk that users actively manage rather than an inevitable or system-wide failure. The findings position AI as an active but culturally constrained co-encoder whose outputs require human contextualization and verification. The study illustrates and refines Hall's encoding-decoding framework and HMC theory within a non-Western, multilingual context, showing how linguistic hybridity and culturally situated interpretation shape AI-mediated meaning construction.
Zitieren
APA
Mia, M. S., Rafi, A. K., & Hania, H. (2026). AI-Mediated Miscommunication in a Multilingual, High-Context Society: Encoding, Decoding, and Meaning Distortion Among Bangladeshi Youth. Journal of Intelligent Digital Media and Communication. https://doi.org/10.67732/jidmc.v1i2.2
Vancouver
Mia MS, Rafi AK, Hania H. AI-Mediated Miscommunication in a Multilingual, High-Context Society: Encoding, Decoding, and Meaning Distortion Among Bangladeshi Youth. Journal of Intelligent Digital Media and Communication. 2026. doi:10.67732/jidmc.v1i2.2
BibTeX
@article{Mia2026,
author = {Mia, Md. Sipon and Rafi, Abhiaz Khan and Hania, Hajrat},
title = {AI-Mediated Miscommunication in a Multilingual, High-Context Society: Encoding, Decoding, and Meaning Distortion Among Bangladeshi Youth},
journal = {Journal of Intelligent Digital Media and Communication},
year = {2026},
doi = {10.67732/jidmc.v1i2.2}
}
EUVIMED