TITLE:
Theory of Higher Categories as a Metalanguage of Strong Artificial Intelligence
AUTHORS:
Mikhail Gennadievich Shcherbakov
KEYWORDS:
Higher-Category Theory, Strong Artificial Intelligence, Morphism, Functor, Natural Transformation, Neural Network, Mathematical Model, Metalanguage
JOURNAL NAME:
Advances in Artificial Intelligence and Robotics Research,
Vol.2 No.2,
June
2,
2026
ABSTRACT: This article discusses the architecture of a strong artificial intelligence neural network. The topic of strong artificial intelligence is one of the most pressing, and various specialists are working in this area. The research gap stems from the lack of a unified theoretical approach to understanding the model and language of strong artificial intelligence and the systematization of AI development in Russia. The aim of this study is to examine theoretical approaches to the concept of a strong artificial intelligence metalanguage. The scientific novelty of this study lies in the mathematical approach to describing the strong artificial intelligence model. The author’s hypothesis is to describe the strong artificial intelligence model using the language of higher-order theory. The essence of this approach is to describe not objects, but their morphisms, functors, and natural transformations. The author notes that the construction of a strong artificial intelligence neural network is based on two axioms of the theory of higher categories: identity and composition. The author provides examples of using the methodology of higher category theory to describe the biological neural network of the human brain, in which neuron weights represent not quantitative values, but connections and their transformations. In the article, the author draws an analogy between the biological neural network of the human brain and the neural network of a strong artificial intelligence capable of solving creative problems like a human, making associations, and learning from limited data. This study utilizes a systems-methodological approach, a hypothetical-deductive approach, and the methodology of the theory of higher categories. In conclusion, we can conclude that the theory of higher categories, based on the axioms of identity and composition, describing morphisms, functors, and natural transformations, is a universal metalanguage for describing the mathematical model of strong artificial intelligence. In other words, the theory of higher categories is a metalanguage of strong artificial intelligence, which allows one to see structure where others see a set of data, to build abstractions of any level, to transfer knowledge based on the similarity of structures, to reflect on one’s own cognitive processes, and to use structural identities to coordinate the structure of thinking.