WHAT HAPPENS NEXT WILL BE DOCUMENTED BY IMAGES AND TEXT
first performed on December 06, 2020
Artist's Home Studio During Lockdown
performed once in 2020
MIRABELLE JONES
Copenhagen, Denmark
792548322m792548322i792548322r792548322a792548322b792548322e792548322l792548322l792548322e792548322j792548322o792548322n792548322e792548322s792548322@792548322g792548322m792548322a792548322i792548322l792548322.792548322c792548322o792548322m
MirabelleJones.com
WHAT HAPPENS NEXT WILL BE DOCUMENTED BY IMAGES AND TEXT
MIRABELLE JONES
“[Box]” is from the series “What Happens Next Will Be Documented in Images and Text,” a set of endurance performances written by an artificial intelligence (GPT-3) and performed by a human being (Mirabelle Jones). “[Box]” was one of four pieces that were shown at ATA Gallery that were performed in the artist’s home studio during Covid-19 lockdown. Instructions for endurance performances were written using OpenAI’s GPT-3, one of the largest natural language processing models in present-day existence. GPT-3 uses machine learning to produce texts which are difficult to distinguish from human text. It was trained on the largest data set to date including articles, books, blog entries, and texts sourced from the internet. As a result, GPT-3’s words sometimes reflect the opinions of the internet and its human authors including biases, prejudices, and hateful output. This has added to existing concerns about artificial intelligence’s increasing presence in our lives.
The performances were generated in response to the prompt, “An Idea for an Endurance Performance Artwork.” These performances touch upon the relationship between technology and the body, authorship and identity, and artificial intelligence creativity vs. reproduction. These relationships generate significant questions. What makes performance art human? What kinds of performance art might an artificial intelligence generate? What can those performances tell us about ourselves and the impressions of “us” we immortalize online?
The following was output by GPT-3, an AI:
“I want to make a performance where I try to do it while: moving a box without using my arms completely non stop for a long time (60 minutes for example); using my arms only a few times throughout a performance and at irregular intervals; not using weight lifting gloves. This is how I imagine it:
-
Find a box to move, one that is light-weight.
-
Set up a time limit for s performance (60 minutes).
-
Start moving the box.
-
When there is one minute left in the performance, stop it, pick up a banana and eat it, then continue moving a box for one more hour.
-
When there are 10 minutes left in the box moving workout, stop it, pick up a lemon and squeeze the juice from it, then continue for 10 more minutes.
-
When there are 5 minutes left, meditate.
The point is to move in one continuous and slowly paced performance from start to finish non-stop.”