Interactive AI-Generated Virtual Instructors Enhance Learning Motivation and Engagement in Financial Education

AI-generated virtual instructors powered by large language models (LLMs) have the potential to augment online learning by making it more interactive and personalized. In this work, we focus specifically on applying this technology to financial literacy education, a critical life skill that is unfortunately undertaught in most school systems. To investigate this potential, we developed three conditions for delivering financial education: A) a Passive Virtual Instructor, B) an Interactive Virtual Instructor powered by LLMs to enable two-way conversation with the learner, and C) a Personally Selected Interactive Virtual Instructor, allowing learners to choose their preferred AI instructor. In a study with 90 participants aged 18–25, we found that incorporating interactivity significantly improved learning motivation, perceived learning facilitation, engagement, and humanness compared to the passive agent. The ability to personally select an instructor provided little additional benefit for most measures. These results highlight the potential of interactive virtual instructors for enhancing motivational and experiential factors, even if comprehension itself is unaffected.