In the era of digitalization, classical ballet is frequently perceived as an inherently conservative art form, entirely detached from neural network experimentation. This article offers a fundamentally different perspective: the academic canon is examined not as a rigid tradition, but as a historically established, strictly deterministic sign system. The study traces the continuous evolution of methods used to record the algorithmic and generative foundations of ballet—from the "track navigation" of the Baroque era to the analytical notations of the 19th and 20th centuries (E. A. Thleur, V. Stepanov, R. von Laban). Furthermore, the scope of inquiry encompasses the transitional phase of video documentation, the introduction of Motion Capture technology, and the contemporary application of artificial intelligence. Particular emphasis is placed on the specific challenge of translating bodily plasticity into a machine-readable digital format optimized for training neural networks. The methodological framework relies on a complementary synthesis of structural-semiotic analysis, comparative-historical methods, and an interdisciplinary convergence of classical art history with modern information technology paradigms. The scientific novelty of this research lies in reconceptualizing classical dance as an art form that inherently possesses algorithmic and generative properties, making it an ideal subject for AI interaction. The study argues that contemporary diffusion models, trained predominantly on the movement vocabulary of modern dance, are objectively incapable of reproducing the strict stylistic features of the classical canon. In this context, the dance notation system developed by V. Stepanov and based on biomechanical principles demonstrates significant practical potential. The main conclusion of this article is that integrating cutting-edge technologies into classical ballet does not disrupt its tradition. Rather, it serves as its logical continuation in a new historical turn. Consequently, generative neural networks should be viewed not as a threat to academicism, but as an innovative tool for objective verification, the reconstruction of lost choreographies, and the preservation of the classical dance canon in the digital age.
Violetta Vladimirovna Zhirova (Fri,) studied this question.