Evolutionary Computing in Translation Studies Under the Transition from the BANI World to the SHIVA World

Abstract

The article examines the transformation of innovative translation studies under the conditions of the global transition from the BANI world to the crisis SHIVA world. The author proves that traditional approaches and static neural networks demonstrate cognitive fragility in the face of dynamic information changes and destructive environments. As a solution, the integration of evolutionary computations (genetic algorithms, swarm intelligence) into translation theory is proposed. The concept of hybrid evolutionary intelligence (Hybrid Evolutionary AI) is described, where large language models generate linguistic diversity, and evolutionary algorithms provide its directed selection and adaptation without complete retraining. The author concludes that translation studies is transitioning into a cybernetic discipline capable of creating autonomous and existing translation systems.