Штучний інтелект як якісно нова форма техногенності: онтологічний та праксеологічний аспекти

Abstract

The paper examines the «black box» problem in neural networks as a specific form of technogenicity. Unlike classical tools, neural networks introduce qualitative opacity, making standard forms of error detection, control, and accountability impossible. The paper distinguishes the ontological dimension of AI's architecture from the praxeological delegation of cognitive authority to it. The core problem lies in the anthropologization of AI: users intuitively project human-like rationality onto a system that operates on fundamentally different principles. This necessitates a philosophical re-evaluation of responsibility criteria in a technology-driven society.