AUTOMATED WEB APPLICATION TESTING BASED ON ARTIFICIAL INTELLIGENCE

Abstract

The evolution of web technologies and the increasing complexity of digital systems have transformed web application testing into an indispensable component of software quality assurance. Traditional automated testing frameworks – based on scripting and static data – remain effective but face scalability and adaptability challenges. This study hypothesizes that the integration of artificial intelligence (AI), particularly large language models (LLMs) and reinforcement learning, can significantly improve the efficiency and autonomy of testing processes. The paper aims to analyze comparatively traditional and AIassisted methods for functional testing of web applications, using a synthesis of recent academic and industrial research. The analysis identifies the main advantages of AI-based testing, such as rapid test generation, extended coverage, and enhanced adaptability, while highlighting limitations related to transparency and integration within continuous integration and continuous delivery (CI/CD) environments. The findings contribute to a better understanding of intelligent automation in software testing and provide guidance for quality assurance (QA) professionals and researchers toward adopting sustainable AI-driven testing practices.