Ensuring high software quality across multiple product lines and projects remains a significant challenge, particularly when a majority of defects originate during the requirements phase. This paper presents an AI-driven framework designed to enhance software quality within the EP department by improving requirement analysis, collateral tracking, and compliance validation.

The proposed solution introduces AI-powered tools that integrate with Jira and Confluence to automatically analyze requirements for granularity, clarity, atomicity, feasibility, verifiability, completeness, and adherence to industry-standard guidelines. By leveraging large language models (LLMs), the system evaluates requirements against best practices, identifies ambiguities and structural deficiencies, and provides actionable recommendations for restructuring into clear, measurable, and implementation-independent statements.

In addition, the framework includes an AI-based collateral status analyzer that monitors document availability and maturity across development phases using CM status accounting. This enables real-time visibility into project readiness, reduces checklist effort, mitigates release risks, and supports better planning decisions.

Initial implementation has demonstrated measurable efficiency improvements in SQA reviews and collateral tracking. The paper also outlines planned enhancements, including AI-driven audit checklist compliance, static code analysis report validation (SCAR/MISRA), and cross-document consistency checks.

Overall, the AI SW Quality framework establishes a scalable and efficient approach to strengthening process and product quality, reducing defects at their source, and improving release predictability through intelligent automation.