How to Glitch AI
What follows are a series of perspectives, which i'm calling <VECTORS>, from which we might begin to consider how, as glitch artists, we could approach producing glitch art in this era of "AI", as defined by algorithms called "artificial neural networks" produced through a process referred to as "machine learning".1
Though I will be focusing primarily on images && communicating through text, it’s important to remember that glitch art can be experienced in multiple dimensions. Glitch art is an interdisciplinary + multi-modal artistic practice, genre, tag, scene, community && aesthetic. Diff artists have diff ways of defining the practice as well as its central concept. My fav definition of a g̴̤̀l̵̹̓i̶̤̍t̴̫̕c̵̫̃ḥ̸̕ is: an unexpected moment in a system that, by catching us off-guard, reveals aspects of that system that might otherwise have gone unnoticed. At its best, a glitch is an epiphany.
My goal w/these <VECTORS> is to suggest ways we might set ourselves up for revelatory surprise. U might find some of the <vector-views> (the image which accompanies each vector) compelling, but it's in the process of creating glitch art that we develop a visceral understanding of the logic + politics mediating our digital lives, && it's this practice i'm aiming on expanding through these <VECTORS>. These are not prescriptions, there is no "right way" to make glitch art, there are only critical, experimental && playful ways of engaging systems the "wrong way".
As artists, cultural producers or knowledge workers of any kind, ignoring the cultural implications of AI means risking irrelevance. It can be hard to keep up w/the fast pace of technological "innovation", but setting aside some time for a bit of "disnovation"0 is as effective (maybe better from this vector-view) as any other way of getting "there". I've suggested only a few <VECTORS> the specifics of which are less important than the general perspective ++ there are certainly many more <VECTORS> to be discovered by anyone willing to explore it actively. [i’m accepting pull requests]
../n!ck.br¡z (2023-2025)