Author(s)
Dhruv Ponda, Dharmik Vaja, Shubham Kanani, Vishakha Savani, Prof. Dhaval R. Chandarana
- Manuscript ID: 140073
- Volume: 2
- Issue: 1
- Pages: 352–357
Subject Area: Computer Science
Abstract
The integration of Artificial Intelligence (AI) into software testing has revolutionized quality assurance, addressing the challenges posed by the increasing complexity of modern software systems. This systematic review synthesizes findings from 45 peer-reviewed studies (2021–2025) to analyze AI-powered test automation tools, their techniques, real-world applications, challenges, and future directions. We examine tools like Testim, Applitools, Mabl, and Functionize, highlighting their capabilities and limitations. Key trends include machine learning (ML) for test case generation, deep learning for defect prediction, and natural language processing (NLP) for automated test creation. While AI enhances efficiency and coverage, challenges such as model interpretability, data quality, test reliability, and adaptation to rapid software changes persist. Recommendations for future research focus on robust AI models, standardized evaluation metrics, and better integration frameworks.