Creative technologies have come a long way in supporting human creativity in a variety of different ways. Perhaps one of the most influential Creativity Support Tools (CST) is Adobe’s Photoshop, which has become a staple in many creative vocations. However, with all its capabilities and features, Photoshop, and other programs like it, never intentionally contribute to an artwork. Modern CSTs have been extremely effective at helping users build up skills, explore creative possibilities, perform complex simulations, and record and track their ideas. Now, with recent advances in artificial intelligence and computational creativity, researchers are beginning to develop co-creative computer applications to enrich the human creative process through collaboration. Instead of just supporting human creativity, this new breed of machine would be able to perceive, analyze, and intentionally contribute to a shared creative artifact. What happens when computers start to become colleagues in the creative process? How can autonomous agents be designed to co-create with humans in open ended and inherently ambiguous domains, such as art? With the Drawing Apprentice, we explore how computers can co-create with humans in abstract art.
The Drawing Apprentice is an enactive AI drawing partner. It is ‘enactive’ in the sense that its contributions are improvisational and based on the input of the user. There is a reciprocal feedback loop between the user and the system. The system models the user’s creative process and artistic style to determine how to collaborate through real time improvisation. As a result, the user is influenced, and perhaps even inspired, by contributions made by the Drawing Apprentice. Instead of a creative human or creative computer, we conceptualize the human-AI pair as a creative system whereby creativity emerges through an interactive and negotiated process of collaboration, experimentation, play, and learning by doing.
Davis, N. Human-Computer Creativity: Blending Human and Computational Creativity. In Proc. of 2013 Conference on Artificial Intelligence and Interactive Digital Entertainment (AIIDE ’13)