An Epistemological Misalignment of Cogs in the AI-Art-Making Machine

Artist, designer, engineer, programmer, technologist and creator are all terms used to describe various roles within the new media art community. It would be easy to assume that there is consensus between these collaborators on the understanding and use of technology. Yet the reality may be quite different - especially in the case of artificial intelligence (AI).

Historically, individuals in the creative disciplines have embraced AI since its inception into both practice and research to produce new media art. In recent times, however, artworks typically employ only a subset of AI technologies, namely machine learning (ML) and deep learning (DL). There may also be differing opinions on what to prioritize: developing a new algorithm, a novel process or producing an aesthetically remarkable output.

The authors first encountered these epistemological challenges as studio-trained interdisciplinary artists and practice-based researchers beginning to integrate AI into their research. Issues discussed here include the challenges of designating universally or unequivocally accepted definitions, the cultural contexts within which technology exists, and how “processed” a technology becomes as a tool can change the language being used.