Towards a Philosophy of Post-creative Practices? – Reading Obvious' “Portrait of Edmond de Belamy”

Abstract

As an emerging experimental subfield of AI in general, artificial creativity-that is, acts of creativity performed (semi-)autonomously by algorithmic/software robots-poses a particular set of problems. Most notably, computational/artificial creativity conflicts with the anthropocentric ways in which we have historically invented 'creativity' as something uniquely and quintessentially human; hence the term 'post-creative'. Yet, when seeking to replicate the kinds of activities (or products) that we are prone to label 'creative', we often tend to forget the contingencies of the 'creativity dispositif' (Reckwitz 2017) and its contested and conflictual character. This amnesia includes the ways in which labelling something as 'creative/not' also, perhaps even primarily, is an aesthetic judgement-either in the traditional sense of philosophical aesthetics, or in the new (Ngai 2012)-rather than merely an ontological statement. To this end, the paper discusses how theories of how art comes into being (sociology of art, the institutional as well as the anti-essentialist theories of art) might have relevance to the issue of creativity as well. Using the recent, heavily debated auctioning of the AI generated painting, "Portrait of Edmond de Belamy" (by Obvious, 2018) as a case this paper will discuss how research into simulating creativity as a productive human activity will have to address not only the new challenges this phenomenon poses, but also some of the older aporias that have long marked our theoretical dealings with the concept 'creativity'.

Understanding and Creating Art with AI: Review and Outlook

“…Epstein et al [43] use the "Portrait of Edmond Belamy" case to explore how anthropomorphization of an AI system influences the perception of humans involved in the creation process. Stephensen [117] discusses the implication that the Belamy case has on the philosophical understanding of creativity. Colton et al [28] discuss how human understanding of different notions of authenticity can be used to address computational authenticity.…”