AI-GENERATED DANCE MOVEMENTS AND CREATIVE OWNERSHIP

Abstract

Artificial intelligence and dance choreography have started to create a paradigm shift in the debate about creativity, authorship, and ownership in digital art. An example of this can be found in AI generated dance works, created using generative models like Recurrent Neural Networks (RNNs), Generative Adversarial Networks (GANs) and even motion capture data generation where machines are able to recreate and innovate in areas of human art. These systems decompose their temporal motion patterns, spatial pathways as well as gestures of expression in order to creatively generate new choreographic sequences independently, posing complicated questions on the authorship of creativity. Conventional models of intellectual property presuppose human-based creativity, but AI generation is the result of the algorithmic combination, not will. The rights of such AI-generated movements are unclear between the choreographer who trained such a model, the creators of the algorithm, or is in the open sphere. Moreover, the assimilation of AI technologies questions aesthetics and ethics, which triggers the redefinition of artistic identity and joint authorship of human beings and machines. The paper will critically analyze the philosophical, legal, and technical aspects of AI-based choreography, focusing on the necessity of new legal provisions and ethical standards, in accordance with the new creative paradigm of hybrid creativity. Through addressing the creative opportunities and the issues of ownership of AI-generated dance, the study helps to extend the discussion concerning the cultural production, the co-creation of human beings and machines, and the changing sense of originality in the era of artificial intelligence.