GENERATIVE ARTIFICIAL INTELLIGENCE FOR AUTOMATED CODE GENERATION, SOFTWARE MAINTENANCE, AND PROGRAMMING EFFICIENCY ENHANCEMENT

Abstract

Generative Artificial Intelligence (GenAI) is rapidly changing software engineering by enabling developers to generate source code, explain programming concepts, identify defects, refactor legacy systems, produce documentation, and automate repetitive maintenance activities. Although experimental studies increasingly report productivity gains from AI-assisted programming, questions remain concerning code reliability, security, verification effort, maintainability, and the extent to which productivity improvements observed in controlled environments transfer to industrial software development. This study examines the perceived and operational impact of GenAI on automated code generation, software maintenance, and programming efficiency through an industrial survey framework. A simulated survey dataset representing 300 software professionals working across software houses, fintech, telecommunications, e-commerce, enterprise software, and health-technology organizations was developed for manuscript demonstration. The proposed instrument measured GenAI adoption, automated code-generation use, maintenance assistance, perceived programming efficiency, code quality, verification burden, security concerns, and organizational governance. Descriptive statistics, correlation analysis, and multiple regression were used to examine relationships among the variables. The simulated findings indicate high adoption of GenAI-assisted development, with code generation, debugging, refactoring, documentation, test generation, and legacy-code explanation emerging as major applications. Programming efficiency showed a positive association with frequency of GenAI use, while perceived code quality and maintenance effectiveness were also positively related to structured human-AI collaboration. However, security concerns, inaccurate outputs, excessive verification requirements, and inadequate organizational policies were identified as important barriers. The findings support the view that GenAI functions more effectively as an augmentation technology than as a replacement for professional software engineering judgment. The study concludes that organizations can obtain greater benefits when AI-assisted programming is integrated with code review, automated testing, security scanning, developer training, and governance mechanisms.