Leveraging Large Language Models in User eXperience (UX) Education when Real Users Are Unavailable

In User eXperience (UX) design, engaging with real users is mandatory and essential, especially in healthcare, where understanding patient and professional experiences is crucial. Yet, ethical and logistical barriers often restrict early-stage access. This study presents a pragmatic approach piloted in a UX workshop for undergraduate engineering students at the University of Western Macedonia, where a Large Language Model (ChatGPT) was used to simulate user perspectives when real participants were unavailable. Students created prompts to emulate patient or clinician responses, using the outputs to develop proto-personas, empathy maps, and design ideas. While LLMs cannot replace authentic user research and may reproduce biases, their thoughtful use can enhance UX education by providing an accessible environment for practicing human-centered methods and preparing students for real-world engagement, particularly in healthcare contexts where user access is limited.