Convergence of AI for Secure Software Development

The traditional approach to software security often fails to detect vulnerabilities early enough. This research explores how artificial intelligence (AI) can be integrated into the software development cycle to improve the identification of vulnerabilities and antipatterns. In the context of DevSecOps, we propose a new methodology that detects antipatterns automatically using machine learning and natural language processing techniques. Also, we explore the possibility of using a large language model (LLM) in a CI (Continuous Integration) and CD (Continuous Delivery) pipeline, to allow the developers real-time feedback, with it the initial phases of the software development cycle can improve the security and quality of the software. Further study is necessary to determine the AI model's true capability in managing real-world challenges.