Un quadro etico e retorico per analizzare i testi generati da LLM
DOI:
https://doi.org/10.60923/issn.2532-8816/22058Parole chiave:
LLM, retorica, figure retiriche, IA, eticaAbstract
La rapida diffusione dei Large Language Model (LLM) sta trasformando la produzione e la circolazione dei contenuti testuali, sollevando importanti questioni linguistiche, cognitive ed etiche. La crescente diffusione di testi generati dall'IA negli ambienti digitali rende necessario comprenderne le caratteristiche retoriche e stilistiche, per valutare l'influenza sui lettori e sull'ecosistema informativo. Questo articolo propone un framework etico e retorico per l'analisi dei testi generati da LLM, fondato su un approccio centrato sul lettore. Il framework identifica un insieme di funzioni retoriche (attendibilità, apologia, prossimità, diversificazione, ambiguità, enfasi, esplicazione, poetica, equità, struttura) e le collega a specifiche figure retoriche, fornendo un metodo strutturato per valutare come le strategie linguistiche plasmino le percezioni e le risposte dei lettori. Sul piano metodologico, lo studio integra la modellazione teorica con il close reading qualitativo, combinando la costruzione deduttiva del framework con l'analisi testuale induttiva. Il framework è testato attraverso l'analisi di due corpora di testi generati da LLM, composti rispettivamente da saggi argomentativi e racconti brevi. I risultati evidenziano schemi retorici ricorrenti, tra cui la sovrarappresentazione delle funzioni enfasi e struttura nei testi argomentativi e la prevalenza del linguaggio figurativo con limitata profondità semantica in quelli narrativi. Questi risultati suggeriscono che gli LLM tendono a riprodurre strutture retoriche ricorrenti, facendo affidamento su strategie di persuasione formale piuttosto che sulla complessità concettuale. Il framework proposto è concepito come un modello aperto, scalabile e integrabile, che può essere progressivamente affinato ed esteso attraverso ulteriori applicazioni empiriche su diverse tipologie testuali e contesti comunicativi.
Riferimenti bibliografici
Alkhrisheh, Hazim Taisir Dayij, Failasofah Failasofah, Taisir Alkhrisheh. 2021. “The Interrelationship among Language Use, Linguistic Competence, and Higher Order Skills.” Indonesian Research Journal in Education 5, no. (april 2021): 6–22. https://doi.org/10.22437/irje.v5i1.11333
Calabrese, Stefano. 2012. Retorica e scienze neurocognitive. Roma: Carocci.
Carrasco-Farre, Carlos. 2024. “Large Language Models are as persuasive as humans, but how? About the cognitive effort and moral-emotional language of LLM arguments.” arXiv:2404.09329. https://doi.org/10.48550/arXiv.2404.09329
Chandra, Rohitash, Abhishek Tiwari, Naman Jain, and Sushrut Badhe. 2024. “Large Language Models for Metaphor Detection: Bhagavad Gita and Sermon on the Mount.” IEEE Access 12: 84452–84469. https://doi.org/10.1109/ACCESS.2024.3411060
Chen, Zhiyi, Jinyi Ye, Emilio Ferrara, and Luca Luceri. 2025. “Prevalence, Sharing Patterns, and Spreaders of Multimodal AI-Generated Content on X during the 2024 U.S. Presidential Election.” arXiv:2502.11248. https://doi.org/10.48550/arXiv.2502.11248
Di Marco, Chrysanne, and Olga Gladkova. 2011. “The RhetFig Project: Computational Rhetorics and Models of Persuasion.” CMNA (Computational Models of Natural Argument) Workshop, San Francisco, CA, August 7–11.
Fish, Stanley. 1980. Is There a Text in This Class? The Authority of Interpretive Communities. Cambridge: Harvard University Press.
Fröhling, Leon, and Arkaitz Zubiaga. 2021. “Feature-Based Detection of Automated Language Models: Tackling GPT-2, GPT-3 and Grover.” PeerJ Computer Science 7. https://doi.org/10.7717/peerj-cs.443
Gadamer, Hans-Georg. 1975. Truth and Method. London: Sheed and Ward.
Giulianelli, Mario, Joris Baan, Wilker Aziz, Raquel Fernández, and Barbara Plank. 2023. “What Comes Next? Evaluating Uncertainty in Neural Text Generators against Human Production Variability.” In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, 14349–371.
Guo, Yanzhu, Guokan Shang, Michalis Vazirgiannis, and Chloé Clavel. 2024. “The Curious Decline of Linguistic Diversity: Training Language Models on Synthetic Text.” In Findings of the Association for Computational Linguistics: NAACL 2024, 3589–3604. Mexico City, Mexico: Association for Computational Linguistics. 10.18653/v1/2024.findings-naacl.228
Hamilton, Craig A., and Ralf Schneider. 2002. “From Iser to Turner and Beyond: Reception Theory Meets Cognitive Criticism.” Style 36, no. 4: 640–58. http://www.jstor.org/stable/10.5325/style.36.4.640
Hanley, Hans W. A., and Zakir Durumeric. 2024. “Machine-Made Media: Monitoring the Mobilization of Machine-Generated Articles on Misinformation and Mainstream News Websites.” arXiv preprint arXiv:2305.09820. https://arxiv.org/abs/2305.09820
Harris, Randy Allen, Chrysanne Di Marco, Ashley Rose Mehlenbacheret et al. 2017. “A cognitive ontology of rhetorical figures.” Cognition and Ontologies: 18–21.
Hemment, Drew, Cody Kommers, et al. 2025. Doing AI Differently: Rethinking the Foundations of AI via the Humanities. White Paper. London: The Alan Turing Institute. https://www.turing.ac.uk/news/publications/doing-ai-differently
Ichien, Nicholas, Dušan Stamenković, and Keith J. Holyoak. 2024. “Large Language Model Displays Emergent Ability to Interpret Novel Literary Metaphors.” Metaphor and Symbol 39, no. 4 (2024): 296–309. https://doi.org/10.1080/10926488.2024.2380348
Jack, Jordynn, and L. Gregory Appelbaum. 2010. “‘This Is Your Brain on Rhetoric’: Research Directions for Neurorhetorics.” Rhetoric Society Quarterly 40 (5): 411–37. doi:10.1080/02773945.2010.516303.
Jakobson, Roman. 1960. “Linguistics and Poetics.” In Style in Language, edited by Thomas A. Sebeok, 350–77. Cambridge, MA: MIT Press.
Kelly, Ashley R., Nike A. Abbott, Randy Allen Harris, Chrysanne DiMarco, and David R. Cheriton. 2010. “Toward an Ontology of Rhetorical Figures.” In Proceedings of the 28th ACM International Conference on Design of Communication (SIGDOC ‘10), 123–30. New York, NY, USA: Association for Computing Machinery. https://doi.org/10.1145/1878450.1878471.
Kendall, Graham, and Teixeira da Silva Jaime A. 2024. “Risks of Abuse of Large Language Models, Like ChatGPT, in Scientific Publishing: Authorship, Predatory Publishing, and Paper Mills.” Learned Publishing 37 (1): 55–62. https://doi.org/10.1002/leap.1578
Kühn, Ramona, Jelena Mitrović, and Michael Granitzer. 2022. "GRhOOT: Ontology of Rhetorical Figures in German." In Proceedings of the Thirteenth Language Resources and Evaluation Conference,, 4001–4010. Marseille, France: European Language Resources Association,. https://aclanthology.org/2022.lrec-1.426/.
Kühn, Ramona, Jelena Mitrović, and Michael Granitzer. 2023. “ESTHER: Ontology of Rhetorical Figures in English.” In Proceedings of the Joint Ontology Workshops 2023 Episode IX: The Quebec Summer of Ontology co-located with the 13th International Conference on Formal Ontology in Information Systems (FOIS 2023), edited by Fumiaki Toyoshima et al., Vol. 3637. CEUR Workshop Proceedings. Sherbrooke, Québec, Canada, July 19–20, 2023. Aachen: CEUR-WS.org.
Lanham, Richard A. 1991. A Handlist of Rhetorical Terms. Berkeley: University of California Press. https://doi.org/10.1525/9780520912045.
Liang, Weixin, Yaohui Zhang, Mihai Codreanu, Jiayu Wang, Hancheng Cao, and James Zou. 2025. “The Widespread Adoption of Large Language Model-Assisted Writing Across Society.” https://arxiv.org/abs/2502.09747
Lucy, Li, and David Bamman. 2021. "Gender and Representation Bias in GPT-3 Generated Stories." In Proceedings of the Third Workshop on Narrative Understanding, edited by Nader Akoury, Faeze Brahman, Snigdha Chaturvedi, Elizabeth Clark, Mohit Iyyer, and Lara J. Martin, 48–55. Virtual: Association for Computational Linguistics. https://doi.org/10.18653/v1/2021.nuse-1.5.
Martínez, Gonzalo, José Alberto Hernández, Javier Conde, Pedro Reviriego, and Elena Merino-Gómez. 2024. “Beware of Words: Evaluating the Lexical Diversity of Conversational LLMs Using ChatGPT as Case Study.” ACM Transactions on Intelligent Systems and Technology. https://doi.org/10.1145/3696459
McKee, Heidi, and James Porter. 2020. “Ethics for AI Writing. The Importance of Rhetorical Context.” In Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society, 110–116. https://doi.org/10.1145/3375627.3375811
Mladenović, Miljana, and Jelena Mitrović. 2013. “Ontology of Rhetorical Figures for Serbian.” In Text, Speech, and Dialogue, edited by Václav Matoušek and Ivan Habernal, 386–93. Lecture Notes in Computer Science. Berlin, Heidelberg: Springer.
Muñoz-Ortiz, Alberto, Carlos Gómez-Rodríguez, and David Vilares. 2024. “Contrasting Linguistic Patterns in Human and LLM-Generated News Text.” Artificial Intelligence Review 57: 265. https://doi.org/10.1007/s10462-024-10903-2
Muzumdar, Prathamesh, Sumanth Cheemalapati, Srikanth Reddy RamiReddy, et al. 2025. “The Dead Internet Theory: A Survey on Artificial Interactions and the Future of Social Media.” Asian Journal of Research in Computer Science 18, no. 1 (January): 67–73. https://doi.org/10.9734/ajrcos/2025/v18i1549
Pilkington, Adrian. 2000. Poetic Effects: A relevance theory perspective. John Benjamins Publishing Company. https://doi.org/10.1075/pbns.75
Raffini, Daniel, Agnese Macori, Lorenzo Porcaro, Tiziana Catarci, and Marco Angelini. 2025a. "How Persuasive Could LLMs Be? A First Study Combining Linguistic-Rhetorical Analysis and User Experiments." 20th International Conference on Artificial Intelligence and Law (ICAIL) LCIC-CLAIRvoyantS Workshop, 2025. arXiv preprint arXiv:2508.09614. https://arxiv.org/abs/2508.09614
Raffini, Daniel, Agnese Macori, Marco Angelini and Tiziana Catarci. 2025b. “A Close Reading Approach to Gender Narrative Biases in AI-Generated Stories.” 2025 IEEE International Conference on Cyber Humanities (IEEE-CH), Florence, Italy, 2025, 1-9. https://doi.org/10.48550/arXiv.2508.09651.
Reinhart, Alex, David West Brown, Ben Markey, et al. 2025. “Do LLMs Write like Humans? Variation in Grammatical and Rhetorical Styles.” Proceedings of the National Academy of Sciences 122, no. 8 (February). https://doi.org/10.1073/pnas.2422455122
Shaib, Chantal, Joe Barrow, Jiuding Sun Sun, Alexa Siu, Byron C. Wallace, and Ani Nenkova. 2024. “Standardizing the Measurement of Text Diversity: A Tool and a Comparative Analysis of Scores.” arXiv:2403.00553v2.
https://doi.org/10.48550/arXiv.2403.00553
Soundararajan, Shweta and Sarah Jane Delany. 2024. "Investigating Gender Bias in Large Language Models Through Text Generation". In Proceedings of the 7th International Conference on Natural Language and Speech Processing (ICNLSP 2024), edited by Mourad Abbas e Abed Alhakim Freihat, 410–424. Trento: Association for Computational Linguistics.
Sun, Zhen, Zongmin Zhang, Xinyue et al. 2025. “Are We in the AI-Generated Text World Already? Quantifying and Monitoring AIGT on Social Media.” arXiv preprint arXiv:2412.18148. https://arxiv.org/abs/2412.18148
Tong, Xiaoyu, Rochelle Choenni, Martha Lewis, and Ekaterina Shutova. 2024. “Metaphor Understanding Challenge Dataset for LLMs.” In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 3517–3536. Bangkok, Thailand: Association for Computational Linguistics
UNESCO. 2021. Recommendation on the Ethics of Artificial Intelligence. United Nations Educational, Scientific and Cultural Organization
Utsumi, Akira. 2005. “The Role of Feature Emergence in Metaphor Appreciation,” Metaphor and Symbol 20, no. 3 (May): 151–72. https://doi.org/10.1207/s15327868ms2003_1
Wang, Yetian, Randy Allen Harris, and Daniel M. Berry. 2021. “An Ontology for Ploke: Rhetorical Figures of Lexical Repetitions.” In Proceedings of the Joint Ontology Workshops 2021 Episode VII: The Bolzano Summer of Knowledge co-located with the 12th International Conference on Formal Ontology in Information Systems (FOIS 2021), and the 12th International Conference on Biomedical Ontologies (ICBO 2021), edited by Emilio M. Sanfilippo et al., Vol. 2969. CEUR Workshop Proceedings. Bolzano, Italy, September 11-18, 2021. Aachen: CEUR-WS.org. http://ceur-ws.org/Vol-2969/paper42-CAOS.pdf.
Werlich, Egon. 1976. A Text Grammar of English. Heidelberg: Quelle & Meyer.
Yakura, Hiromu, Lopez-Lopez Ezequiel, Brinkmann Levin, et al. 2025. “Empirical evidence of Large Language Model's influence on human spoken communication”. arXiv:2409.01754
Yao, Ben, Yazhou Zhang, Qiuchi Li, and Jing Qin. 2024 “Is Sarcasm Detection A Step-by-Step Reasoning Process in Large Language Models?” arXiv:2407.12725. https://doi.org/10.48550/arXiv.2407.12725
Yoo, Dahey, Hyunmin Kang, and Changhoon Oh. 2025. “Deciphering Deception: How Different Rhetoric of AI Language Impacts Users’ Sense of Truth in LLMs.” International Journal of Human–Computer Interaction 41, no. 4: 2163–2183. https://doi.org/10.1080/104
Zhao, Xia, and Xiliang Cui. “Rhetorical Studies in the Context of Large Language Models (LLMs).” Apollo - University of Cambridge Repository, 2024. https://doi.org/10.17863/CAM.108500
Downloads
Pubblicato
Come citare
Fascicolo
Sezione
Licenza
Copyright (c) 2026 Daniel Raffini, Agnese Macori, Marco Angelini, Tiziana Catarci
Questo lavoro è fornito con la licenza Creative Commons Attribuzione 4.0 Internazionale.
