What happened to the subject? Mediated anticipation in neural painting

This paper presents a phenomenology of artistic painting as an anticipatory process. I propose that the artist seeks to establish a state of equilibrium in a model of self-awareness expressed and represented in a self-constituted physical artefact intended to communicate to others, not representationally but affectively. ‘Neural painting’ is an arts-based research method employing a simple computational model of human aesthetic discrimination to study the creative realization of the artistic image. I use this method to explore the relationship of self and ‘other’ in computationally mediated self-portraiture. I develop an image in an exchange with a neural network by reflecting on its output and inputting autographic modifications to those images, blending visceral gesture with the ‘black box’ of artificial intelligence (AI). Through this deeply personalized and perhaps agonistic interchange between organic self and algorithmic reflection, I seek to expose the tacit mediation implicit in the technical artefact, opening an understanding of the existential relations between natural systems (the artist) and technical entities positioned as collaborators in an anticipatory aesthetics.