What are you looking for today?

gilchrist, Bruce (2022) Poetics of Artificial Intelligence in Art Practice: (Mis)apprehended Bodies Remixed as Language. Doctoral thesis, University of Sunderland.

Document

Abstract

Abstract

With a focus on the last five years, art employing artificial intelligence (AI) has been defined by a spectrum of activity, from the deep learning explorations of neural network researchers to artists critiquing the broader social implications of AI technology. There is an emergence of and increasing access to new tools and techniques for repurposing and manipulating material in unprecedented ways in art.
At the same time, there is a dearth of language outside the scientific domain with which to discuss it. A combination of contextual review, comparison of artistic approaches, and practical projects explores speculation that the conceptual
repertoire for remix studies can open up to art enabled by AI and machine learning (ML).
This research contributes a practical, conceptual and combinatorial approach for artists who do not necessarily have a grounding in engineering or computer science. A bricolage methodology—described by Annette Markham as combining serendipity, proximity and contingency—reveals the poetics of AI-enabled art in the form of an assemblage of techniques that understands poetics as active making
(poiesis) as well as an approach to manipulating language.
The poetic capacity of AI/ML is understood as an emergent form of remix technique, with the ML at its core functioning like a remix engine. This practice based research presents several projects founded on an interrelation of body, text
Bruce Gilchrist. Poetics of Artificial Intelligence in Art Practice: (Mis)apprehended Bodies Remixed as Language. 3 and predictive technology enabled by a human-action-recognition algorithm combined with a natural language generator. A significant number of artistic works have been made around object and facial recognition, while very little (if any) artist
activity has focused on human-action-recognition. For this reason, I concentrated my research there.

Item Type: Thesis (Doctoral)
Depositing User: Veronique Laniel
URI: https://sure.sunderland.ac.uk/id/eprint/14711

Further Details

More Information

Depositing User

Identifiers

Item ID
14711

Users with ORCIDS

Catalogue record

Date Deposited
04 Apr 2022 13:52
Last Modified
03 Jun 2025 14:03

Contributors

Author
Bruce gilchrist

University Divisions

Subjects

Actions (login required)

View Item (Repository Staff Only)

Downloads per month over past year