Home > Research > Publications & Outputs > AI, be less ‘stereotypical’

Electronic data

  • 3_Tantucci Sparvoli_MN_5-28-2025

    Accepted author manuscript, 3.24 MB, Word document

    Available under license: CC BY: Creative Commons Attribution 4.0 International License

Links

Text available via DOI:

View graph of relations

AI, be less ‘stereotypical’: ChatGPT’s speech is conventional but never unique

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

AI, be less ‘stereotypical’: ChatGPT’s speech is conventional but never unique. / Tantucci, Vittorio; Sparvoli, Cralotta.
In: Intercultural Pragmatics, Vol. 22, No. 2, 04.08.2025, p. 231-258.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

APA

Vancouver

Tantucci V, Sparvoli C. AI, be less ‘stereotypical’: ChatGPT’s speech is conventional but never unique. Intercultural Pragmatics. 2025 Aug 4;22(2):231-258. doi: 10.1515/ip-2025-2003

Author

Tantucci, Vittorio ; Sparvoli, Cralotta. / AI, be less ‘stereotypical’ : ChatGPT’s speech is conventional but never unique. In: Intercultural Pragmatics. 2025 ; Vol. 22, No. 2. pp. 231-258.

Bibtex

@article{04d7d1ade31f4462b6da55d443969ff7,
title = "AI, be less {\textquoteleft}stereotypical{\textquoteright}: ChatGPT{\textquoteright}s speech is conventional but never unique",
abstract = "Can AI reproduce human interaction? It can, but only stereotypically. While it can simulate (and even exaggerate) dialogic engagement, its lexicon is less diverse, and the speech acts it realizes are more repetitive and less varied (we took directives as an example). Most importantly, AI struggles to represent {\textquoteleft}conversational uniqueness{\textquoteright}, that is ways to interact that define the specificity of a particular conversation and are not entirely conventional. We discovered this after analyzing dialogic resonance (the re-use of an interlocutor{\textquoteright}s construction), recombinant creativity (the creative reformulation of an interlocutor{\textquoteright}s construction), relevance acknowledgement (the acknowledgement of what an interlocutor said) and other variables from a sampled section of the CallHome Corpus of Chinese telephone conversations. After feeding ChatGPT with speakers{\textquoteright} demographics and contextual information, we asked it to reproduce telephone interactions among Chinese family members. We then fitted a conditional mixed effects Bayesian regression comparing the two datasets. We found that AI over-generalizes human dialogue. It provides a stereotypical way of conversing but shows scarce flexibility in including {\textquoteleft}atypical{\textquoteright} and unconventional utterances, which are, in turn, constitutive of human interactions that occur in real life.",
keywords = "AI, Pragmatics, computational linguistics, alignment, resonance",
author = "Vittorio Tantucci and Cralotta Sparvoli",
year = "2025",
month = aug,
day = "4",
doi = "10.1515/ip-2025-2003",
language = "English",
volume = "22",
pages = "231--258",
journal = "Intercultural Pragmatics",
issn = "1612-295X",
publisher = "Walter de Gruyter GmbH & Co. KG",
number = "2",

}

RIS

TY - JOUR

T1 - AI, be less ‘stereotypical’

T2 - ChatGPT’s speech is conventional but never unique

AU - Tantucci, Vittorio

AU - Sparvoli, Cralotta

PY - 2025/8/4

Y1 - 2025/8/4

N2 - Can AI reproduce human interaction? It can, but only stereotypically. While it can simulate (and even exaggerate) dialogic engagement, its lexicon is less diverse, and the speech acts it realizes are more repetitive and less varied (we took directives as an example). Most importantly, AI struggles to represent ‘conversational uniqueness’, that is ways to interact that define the specificity of a particular conversation and are not entirely conventional. We discovered this after analyzing dialogic resonance (the re-use of an interlocutor’s construction), recombinant creativity (the creative reformulation of an interlocutor’s construction), relevance acknowledgement (the acknowledgement of what an interlocutor said) and other variables from a sampled section of the CallHome Corpus of Chinese telephone conversations. After feeding ChatGPT with speakers’ demographics and contextual information, we asked it to reproduce telephone interactions among Chinese family members. We then fitted a conditional mixed effects Bayesian regression comparing the two datasets. We found that AI over-generalizes human dialogue. It provides a stereotypical way of conversing but shows scarce flexibility in including ‘atypical’ and unconventional utterances, which are, in turn, constitutive of human interactions that occur in real life.

AB - Can AI reproduce human interaction? It can, but only stereotypically. While it can simulate (and even exaggerate) dialogic engagement, its lexicon is less diverse, and the speech acts it realizes are more repetitive and less varied (we took directives as an example). Most importantly, AI struggles to represent ‘conversational uniqueness’, that is ways to interact that define the specificity of a particular conversation and are not entirely conventional. We discovered this after analyzing dialogic resonance (the re-use of an interlocutor’s construction), recombinant creativity (the creative reformulation of an interlocutor’s construction), relevance acknowledgement (the acknowledgement of what an interlocutor said) and other variables from a sampled section of the CallHome Corpus of Chinese telephone conversations. After feeding ChatGPT with speakers’ demographics and contextual information, we asked it to reproduce telephone interactions among Chinese family members. We then fitted a conditional mixed effects Bayesian regression comparing the two datasets. We found that AI over-generalizes human dialogue. It provides a stereotypical way of conversing but shows scarce flexibility in including ‘atypical’ and unconventional utterances, which are, in turn, constitutive of human interactions that occur in real life.

KW - AI

KW - Pragmatics

KW - computational linguistics

KW - alignment

KW - resonance

U2 - 10.1515/ip-2025-2003

DO - 10.1515/ip-2025-2003

M3 - Journal article

VL - 22

SP - 231

EP - 258

JO - Intercultural Pragmatics

JF - Intercultural Pragmatics

SN - 1612-295X

IS - 2

ER -