The Use of Generative AI Tools by Requirements Engineers: An Interview with Industry Professionals

Abstract

This paper investigates how Generative Artificial Intelligence (GenAI) grounded in Large Language Models (LLMs) are being employed to support activities within Requirements Engineering (RE). Based on interviews conducted with software engineering professionals, the findings indicate that GenAI tools are particularly valuable during requirements elicitation and specification, contributing to improved productivity and reduced operational effort, especially when addressing Functional Requirements (FRs). Despite these benefits, participants also reported important limitations, including sensitivity to prompt formulation, generation of overly generic outputs, and limited capability to adequately support Non-Functional Requirements (NFRs).