Designing With AI : A Study of Non-Developer Software Creation Using ChatGPT

This thesis investigates the potential of enabling non-developers to create functional software applications comparable to those built by experienced developers through the use of AI-assisted coding tools, specifically focusing on ChatGPT. The study explores whether AI-powered code generation can effectively bridge the gap between users with no programming background and professional developers, thus democratizing software development. A mixed-methods approach was employed, including design and prototyping, user studies with participants of varying technical expertise, and qualitative post-task interviews. Key metrics such as task completion time, and user satisfaction were analyzed to assess the quality and efficiency of software built with AI support. Results indicate that while AI assistance significantly lowers the barrier to entry for non-developers, challenges remain in achieving parity with expert developers in complex programming tasks. The findings contribute to understanding the capabilities and limitations of AI coding tools in interaction design and software engineering contexts and suggest avenues for future research in human-AI collaboration for software development.