This thesis examines the ability of narratives produced by generative artificial intelligence tools to elicit emotional responses in readers and to offer forms of identification comparable to those generated by human writing. Starting from a reflection on narration as an anthropological need, a means of identity construction, and a tool for the emotional processing of experience, the research is situated within the field of digital communication and develops a linguistic analysis of textual production. The study focuses on narrative production associated with paraliterature, with particular attention to children’s and young adult fiction. It does so through an overview of the first Italian editorial products created with the support of generative artificial intelligence and published on digital platforms. The experimental section involves the generation of short stories using three large language models and their evaluation by a group of young readers through an anonymous survey. The findings indicate that generative artificial intelligence can produce emotionally engaging narratives, while also reaffirming the central role of the human being in the interpretative, creative, and ethical processes, as well as in the relationship between humans and machines. This thesis investigates the capacity of narratives generated by artificial intelligence systems to evoke emotional responses in readers and to provide forms of affective and interpretive identification comparable to those produced by human authorship. Grounded in a theoretical reflection on narration as an anthropological practice—central to identity formation and to the emotional mediation of experience—the study is situated within the field of digital media and communication and adopts a linguistic and discourse-oriented approach to textual analysis. The research examines narrative production within the domain of paraliterature, with particular emphasis on children’s and young adult fiction. It draws on an analysis of the earliest Italian editorial experiments involving generative artificial intelligence published on digital platforms, alongside a comparative literary experiment contrasting a short story written by an emerging author with texts generated by a system of generative AI applications. The empirical component of the study consists in the generation of narrative texts by three large language models and their subsequent evaluation by a group of young readers through an anonymous survey. The results suggest that generative artificial intelligence can produce emotionally engaging narratives; however, they also underscore the enduring centrality of human agency in interpretive, creative, and ethical processes, as well as in shaping the relationship between humans and machines. Keywords: LLM, Creative Writing, Children’s and Young Adult Literature, and Italian Language Studies
La mia tesi analizza la capacità delle narrazioni prodotte da strumenti d’intelligenza artificiale generativa di suscitare emozioni nei lettori e di offrire forme di rispecchiamento paragonabili a quelle della scrittura umana. Muovendo da una riflessione sulla narrazione come bisogno antropologico, strumento di costruzione dell’identità e di elaborazione emotiva dell’esperienza, la ricerca si colloca nel contesto della comunicazione digitale, approfondendo un’analisi linguistica dei testi. Lo studio prende in esame la produzione narrativa riconducibile alla paraletteratura, con particolare attenzione ai racconti per l’infanzia e per adolescenti, attraverso una ricognizione dei primi prodotti editoriali italiani realizzati con l’ausilio dell’intelligenza artificiale generativa, poi pubblicati su piattaforme digitali. La parte sperimentale comprende la generazione di racconti tramite tre modelli linguistici di grandi dimensioni e la loro valutazione da parte di un gruppo di giovani lettori, attraverso un sondaggio anonimo. I risultati evidenziano come l’intelligenza artificiale generativa possa produrre narrazioni emotivamente coinvolgenti, pur confermando la centralità dell’essere umano nel processo interpretativo, creativo ed etico e nella relazione tra uomo e macchina. Parole chiave: LLM, scrittura creativa, letteratura per ragazzi, lingua italiana.
LA LINGUA DELLA LETTERATURA PRODOTTA DAGLI LLM. UNA RICERCA SPERIMENTALE SU RACCONTI PER ADOLESCENTI.
BELLINGERI, SABRINA
2024/2025
Abstract
This thesis examines the ability of narratives produced by generative artificial intelligence tools to elicit emotional responses in readers and to offer forms of identification comparable to those generated by human writing. Starting from a reflection on narration as an anthropological need, a means of identity construction, and a tool for the emotional processing of experience, the research is situated within the field of digital communication and develops a linguistic analysis of textual production. The study focuses on narrative production associated with paraliterature, with particular attention to children’s and young adult fiction. It does so through an overview of the first Italian editorial products created with the support of generative artificial intelligence and published on digital platforms. The experimental section involves the generation of short stories using three large language models and their evaluation by a group of young readers through an anonymous survey. The findings indicate that generative artificial intelligence can produce emotionally engaging narratives, while also reaffirming the central role of the human being in the interpretative, creative, and ethical processes, as well as in the relationship between humans and machines. This thesis investigates the capacity of narratives generated by artificial intelligence systems to evoke emotional responses in readers and to provide forms of affective and interpretive identification comparable to those produced by human authorship. Grounded in a theoretical reflection on narration as an anthropological practice—central to identity formation and to the emotional mediation of experience—the study is situated within the field of digital media and communication and adopts a linguistic and discourse-oriented approach to textual analysis. The research examines narrative production within the domain of paraliterature, with particular emphasis on children’s and young adult fiction. It draws on an analysis of the earliest Italian editorial experiments involving generative artificial intelligence published on digital platforms, alongside a comparative literary experiment contrasting a short story written by an emerging author with texts generated by a system of generative AI applications. The empirical component of the study consists in the generation of narrative texts by three large language models and their subsequent evaluation by a group of young readers through an anonymous survey. The results suggest that generative artificial intelligence can produce emotionally engaging narratives; however, they also underscore the enduring centrality of human agency in interpretive, creative, and ethical processes, as well as in shaping the relationship between humans and machines. Keywords: LLM, Creative Writing, Children’s and Young Adult Literature, and Italian Language Studies| File | Dimensione | Formato | |
|---|---|---|---|
|
TESI Sabrina Bellingeri CoD.pdf
accesso aperto
Dimensione
5.08 MB
Formato
Adobe PDF
|
5.08 MB | Adobe PDF | Visualizza/Apri |
È consentito all'utente scaricare e condividere i documenti disponibili a testo pieno in UNITESI UNIPV nel rispetto della licenza Creative Commons del tipo CC BY NC ND.
Per maggiori informazioni e per verifiche sull'eventuale disponibilità del file scrivere a: unitesi@unipv.it.
https://hdl.handle.net/20.500.14239/32902