GENERATIVE ARTIFICIAL INTELLIGENCE AGENTIC SOLUTIONS FOR SOFTWARE QUALITY ASSURANCE

Abstract

AI agents demonstrate autonomy in planning, execution, analysis, and maintenance, offering potential to address
persistent challenges in quality assurance such as maintenance overhead, lengthy debugging cycles, and limited predictive
capacity. To structure the investigation, seven core agentic use cases were defined, covering the full lifecycle of test automation.
A comparative evaluation of three SaaS platforms—KaneAI, Zephyr Scale Automate, and TestRigor—was conducted using a
dual framework that combines quantitative scoring with qualitative assessment. This approach enables an integrated
understanding of both the maturity of support for agentic functionality and the qualitative depth of its implementation,
contributing to ongoing discourse on the evolution of software testing practices.