Development of Quality Assurance Standards for the Industrialization of Software Applications Prototyped Using Intelligent Assistants

Abstract

The article is dedicated to the analysis of quality assurance transformation in the context of industrializing software applications prototyped using intelligent assistants. The relevance of the study is determined by the rapid diffusion of large language models into software development workflows and the growing mismatch between accelerated prototyping and delayed quality degradation in production environments. Scientific novelty lies in the analytical reinterpretation of quality assurance as a layered, distributed regulatory system rather than a terminal verification phase. The work describes structural shifts in assurance logic, including constraint-driven generation, embedded testing, knowledge grounding, and orchestration-level governance. Special attention is paid to the temporal displacement of quality signals and the limitations of metric-centered evaluation in AI-assisted development. The work sets itself the goal of conceptualizing quality assurance standards suitable for the industrialization of AI-prototyped software. To achieve this goal, analytical synthesis and comparative analysis are used. A corpus of recent studies on generative AI, program repair, knowledge-enhanced systems, autonomous agents, and software quality tools is examined. The conclusions demonstrate that no single assurance paradigm stabilizes quality under AI augmentation, and layered standards emerge as a structural necessity. The article will be useful for researchers, software architects, quality engineers, and technology managers.