Leveraging Ai-driven defect prediction models for enhancing software quality assurance

Artificial intelligence has brought change into the software quality assurance domain by applying techniques such as defect prediction, automation, and efficiency in testing. This study reviews AI defect prediction models in the context of software assurance with respect to reliability enhancement and the optimization of testing processes. A whole range of machine learning techniques are reviewed with respect to defect detection, data privacy issues, model interpretability, computational costs, and further challenges. AI in software quality assurance shall then be discussed on the themes of future intelligent defect prediction, incorporation into DevOps and CI/CD pipelines, and the role of explainable AI (XAI) in offering white box feedback. The artificial intelligence activities in the domain of software quality assurance have disregarded conventions and pursued much-needed paths into the truly proactive defect management domain, aiming to reduce manual intervention and provide better quality in software, all this against their own limitations. This study also attempts to feed into the debate concerning AI in SQA by suggesting directions for future studies that might do work towards better model accuracy, interpretability, and performance.