Communication orale non publiée/Abstract (Colloques, congrès, conférences scientifiques et actes)
The future of CHI(Art): Can Body of Text Replace a Real Body?
Gubenko, Alla; TAVARES VANDEN BERGHE, Anastasia
2023ACM CHI’23 Workshop
 

Documents


Texte intégral
Submission The future of CHI(Art)- Can Body of Text Replace a Real Body (2).pdf
Postprint Éditeur (414.54 kB)
Télécharger

Tous les documents dans ORBilu sont protégés par une licence d'utilisation.

Envoyer vers



Détails



Mots-clés :
Text-based co-creation; prompt engineering; text-to-image generation; artificial creative aids; creative deskilling; generative art
Résumé :
[en] Recently a variety of new creativity support tools (e.g., Midjourney, DALL·E 2, Stable Diffusion) has been launched, making the creative process as accessible as ever. However, these new artificial creative aids—Text-to-Image Generation models — ultimately hinge on human textual prompts. Using only a textual description, a person can generate new, high-quality images without previous art training or learning domain-specific skills. The adoption of these novel artistic tools is accompanied by the development of online marketplaces where one can buy successful prompts. The new type of creative process becomes more and more linguistically loaded and disembodied, i.e., not requiring any physical and multimodal interaction with artistic materials, tools, or media. This paper visualizes such disembodied creative practice and triggers reflections on the future of art and the impact of technology on human domain-related skills.
Disciplines :
Psychologie cognitive & théorique
Auteur, co-auteur :
Gubenko, Alla
TAVARES VANDEN BERGHE, Anastasia ;  University of Luxembourg > Faculty of Humanities, Education and Social Sciences (FHSE)
Co-auteurs externes :
no
Langue du document :
Anglais
Titre :
The future of CHI(Art): Can Body of Text Replace a Real Body?
Date de publication/diffusion :
28 avril 2023
Nom de la manifestation :
ACM CHI’23 Workshop
Date de la manifestation :
28-04-2023
Manifestation à portée :
International
Références de l'abstract :
Recently a variety of new creativity support tools (e.g., Midjourney, DALL·E 2, Stable Diffusion) has been launched, making the creative process as accessible as ever. However, these new artificial creative aids—Text-to-Image Generation models — ultimately hinge on human textual prompts. Using only a textual description, a person can generate new, high-quality images without previous art training or learning domain-specific skills. The adoption of these novel artistic tools is accompanied by the development of online marketplaces where one can buy successful prompts. The new type of creative process becomes more and more linguistically loaded and disembodied, i.e., not requiring any physical and multimodal interaction with artistic materials, tools, or media. This paper visualizes such disembodied creative practice and triggers reflections on the future of art and the impact of technology on human domain-related skills.
Disponible sur ORBilu :
depuis le 15 mai 2023

Statistiques


Nombre de vues
135 (dont 6 Unilu)
Nombre de téléchargements
87 (dont 0 Unilu)

Bibliographie


Publications similaires



Contacter ORBilu