A SYSTEMATIC REVIEW AND CONCEPTUAL FRAMEWORK FOR INTEGRATING GENERATIVE AI INTO THE SOFTWARE DEVELOPMENT LIFECYCLE (SDLC)

Abstract

The swift advancement of Generative Artificial Intelligence (GenAI) techniques including large language models, autonomous coding programs, and intent-based software development methodologies has radically transformed software engineering processes across the entire Software Development Lifecycle (SDLC). While industry use of GenAI tools like ChatGPT and GitHub Copilot continues to gain momentum, the research domain is characterized by fragmentation, phase-centricity, and a lack of integrated frameworks considering governance issues. This paper presents a Systematic Literature Review consistent with the PRISMA guidelines, consolidating 71 scholarly articles from IEEE Xplore, ACM Digital Library, SpringerLink, and Elsevier ScienceDirect between 2021 and 2026 to explore the capabilities of GenAI across all SDLC phases. The findings confirm considerable productivity and automation opportunities along with inherent limitations relating to hallucination, spread of security vulnerabilities, algorithmic biases, and knowledge monopoly. Ten important research gaps are uncovered, and an innovative theoretical framework is developed, comprising phase-wise GenAI tool maps with a Human-AI Collaboration Layer and Governance/Ethics Layer