Requirements Engineering (RE) process in software engineering projects is becoming increasingly resource-intensive, making automation essential. The software requirement specification (SRS) document provides significant guidance and defines the final product. These documents are usually written in natural language (NL), which is a convenient wayof representing them. As software projects in industrial systems expand, the software requirement specification documents are also growing, becoming more challenging to analyze and potentially containing ambiguities. With the advent of Generative Artificial Intelligence (GenAI), automating tasks like ambiguity detection is now feasible due to AI’s ability to process and understand various contexts. This thesis explores the use of GenAI in RE to address ambiguities in textual requirements. We propose a structured prompt pattern through prompt engineering technique, designed to optimize the performance of Large Language Models (LLMs) to specifically tackle the ambiguities in textual requirements, and develop a web-based tool that integrates different GPT versions such as GPT-3.5 and GPT4 and offers a user-friendly, chat-like interface for users to interact with the AI. Through continuous feedback, users can identify and resolve ambiguities more effectively. We evaluate the effectiveness of these approaches by applying them to real-world requirements from Alstom and literature-based datasets.