Who is an Artist in AI Era?
Publication
Lev Manovich. AI & Myths of Creativity. (Draft version of Chapter 2 of Artificial Aesthetics by Lev Manovich and Emanuele Arielli)
Abstract
As digital media continues to evolve, the concept of what it means to be an artist periodically undergoes significant change. With artificial intelligence (AI) now capable of producing art that can stand alongside human-created works, we are faced again with rethinking our understanding of art, creativity, and authorship. In this chapter, I examine how AI's growing abilities challenge traditional views, suggesting that we might need a new kind of Turing test—one that evaluates creativity.
For example, for many years, all popular digital creation tools like Photoshop or Blender were already incorporating AI-like features. Since 2013, Photoshop had a "contextual fill" tool that automatically generates an area of an image to match patterns around it, while Microsoft Word has had a function to "summarize" documents. The use of these digital tools has already blurred the lines between human and machine authorship.
The developments of multiple types of art, each with its own criteria of quality and success, further complicate the question, "Who is an artist?" Earlier ways of judging art, such as through formal training or critical praise, now coexist with digital-age measures like social media popularity, where artists can achieve recognition without traditional credentials.
The chapter concludes by suggesting that simply simulating the results of the artistic process with computers is not sufficient. A computer needs to learn the creative process as opposed to just simulating the end product. This approach could lead AI to become a more original creative force, changing its role from that of a tool to that of a genuine collaborator.