An Exploratory Study on Framework-Oriented AI Support for Secure SDLC Practices

Modern software systems are becoming increasingly complex, while security issues continue to grow across different stages of the Software Development Lifecycle (SDLC). In many development environments, security and analytical support are often addressed late in the process, which increases risk and development effort. Although artificial intelligence has shown promise in supporting software engineering activities, most existing approaches focus on specific tasks or individual SDLC phases rather than the entire lifecycle. This paper proposes a conceptual framework for AI-assisted support in the Secure Software Development Lifecycle (Secure SDLC). The framework is developed through a review of existing research studies, industry practices, and secure development standards. Instead of presenting a working system or experimental results, the study focuses on defining high-level framework components and explaining their conceptual roles across SDLC phases such as requirements, design, development, testing, deployment, and maintenance. The proposed framework aims to support security awareness and decision-making without disrupting existing development processes. The main contribution of this work is an implementation-independent framework that connects secure SDLC principles with AI-assisted analytical concepts. This study serves as a foundation for future research, system development, and practical validation in real-world software engineering environments.