This study investigates the application of artificial intelligence to optimize the creation of user acceptance test (UAT) cases using large language models (LLMs) such as advanced AI text generators (e.g., ChatGPT, Claude AI, Gemini, and Copilot). These AI-driven tools, powered by LLMs, efficiently produce test cases, enhancing the reliability and efficiency of the testing process. The research rigorously evaluates these AI-driven approaches through empirical tests and their integration into real-world software testing settings. The performance evaluation focuses on precision, recall, F-score, accuracy, test coverage, average generation time, language proficiency, ease of use, and real-world applicability. The goal is to seamlessly integrate these tools into the standard testing workflows of a software testing and quality assurance company, reducing the workload on human testers by automating test case generation. This automation aims to enhance both the breadth and accuracy of tests and facilitate the creation of complex test scenarios. Experimental results indicate significant advancements in software testing methodologies compared to traditional methods such as Boundary Value Analysis (BVA), Equivalence Partitioning (EP), complex BVA, use case, and state transition testing. AI tools notably improve the formulation and implementation of UATs, saving time and ensuring software solutions align more effectively with user requirements and expectations. Overall, AI-powered text generators, particularly ChatGPT models (versions 4.0 and 3.5), demonstrated strong performance and high user satisfaction in generating UAT test cases, marking a significant step forward in automated software testing.