This research evaluates how Large Language Models (LLMs) perform against human experts in requirements engineering, a critical but understudied area of software development.

Key Findings:

• LLMs show promising capabilities in requirements elicitation, potentially augmenting human expertise
• Current research focuses heavily on code generation, while requirements engineering applications remain underexplored
• Requirements gathering from stakeholders represents a significant opportunity for AI assistance
• The study bridges an important gap between AI advancement and practical software development needs

Why It Matters: Requirements engineering is often the foundation of successful software projects. Understanding how AI can support this process could reduce development costs, improve requirement quality, and accelerate project timelines.

Analysis of LLMs vs Human Experts in Requirements Engineering