Introducing the exhibition Le monde selon l'AI (Paris, Jeu de Paume, April-September 2025), curator Antonio Somaini records the importance of reconsidering the rhetorical and procedural tradition of ekphrasis in order to understand some aspects of generative AI programs. These are, in this case, Text-to-Image (TTI) models in which the interaction starts with a textual trigger, the so-called prompt, which directs the program in the exploration - gradually refined by further descriptions - of immense datasets of images, used to train the program itself to recognize, select, mix, and invent new visual configurations, working in deep, latent areas of its own structure. Comparing ekphrasis and prompt writing (prompt engineering, prompt design) “can shed light on a number of aspects of our experience of textual and visual works, including the role of our imagination, our emotions, and what happens when human agency within these imaginative and emotional processes is replaced by the artificial agency of an AI system” (Verdicchio 2024). Also described as a form of “operative ekphrasis” (Bajohr 2024), the prompt is, in short, at the center of interdisciplinary and creative looks, involving artworks such as the installation Ekphrasis, by Spanish collective Estampa (2025), which applies Image-to-Text models to fragments of cinema to generate detailed descriptions that reveal “the poetic potential inherent in the radical impossibility of translating images into words and vice versa”. The paper traverses this encounter between the ecphrasis and the prompt by presenting some recent works and a conversation with art collective Estampa.

L'incontro tra ecfrasi e Intelligenza Artificiale, con uno sguardo all'arte contemporanea / Sbrilli, Antonella. - In: LA RIVISTA DI ENGRAMMA. - ISSN 1826-901X. - 229:(2025), pp. 161-172.

L'incontro tra ecfrasi e Intelligenza Artificiale, con uno sguardo all'arte contemporanea

Sbrilli, Antonella
2025

Abstract

Introducing the exhibition Le monde selon l'AI (Paris, Jeu de Paume, April-September 2025), curator Antonio Somaini records the importance of reconsidering the rhetorical and procedural tradition of ekphrasis in order to understand some aspects of generative AI programs. These are, in this case, Text-to-Image (TTI) models in which the interaction starts with a textual trigger, the so-called prompt, which directs the program in the exploration - gradually refined by further descriptions - of immense datasets of images, used to train the program itself to recognize, select, mix, and invent new visual configurations, working in deep, latent areas of its own structure. Comparing ekphrasis and prompt writing (prompt engineering, prompt design) “can shed light on a number of aspects of our experience of textual and visual works, including the role of our imagination, our emotions, and what happens when human agency within these imaginative and emotional processes is replaced by the artificial agency of an AI system” (Verdicchio 2024). Also described as a form of “operative ekphrasis” (Bajohr 2024), the prompt is, in short, at the center of interdisciplinary and creative looks, involving artworks such as the installation Ekphrasis, by Spanish collective Estampa (2025), which applies Image-to-Text models to fragments of cinema to generate detailed descriptions that reveal “the poetic potential inherent in the radical impossibility of translating images into words and vice versa”. The paper traverses this encounter between the ecphrasis and the prompt by presenting some recent works and a conversation with art collective Estampa.
2025
ecfrasi; arte contemporanea; text-to-image
01 Pubblicazione su rivista::01a Articolo in rivista
L'incontro tra ecfrasi e Intelligenza Artificiale, con uno sguardo all'arte contemporanea / Sbrilli, Antonella. - In: LA RIVISTA DI ENGRAMMA. - ISSN 1826-901X. - 229:(2025), pp. 161-172.
File allegati a questo prodotto
File Dimensione Formato  
Sbrilli_Ecfrasi_2025.pdf

accesso aperto

Tipologia: Versione editoriale (versione pubblicata con il layout dell'editore)
Licenza: Creative commons
Dimensione 1.74 MB
Formato Adobe PDF
1.74 MB Adobe PDF

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11573/1757416
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact