Description: Qualitative interviewing is a widely used research method that is very effective for gaining insights into the target user group. However, they are often time-consuming, typically result in small sample sizes, which in turn may lack diverse perspectives. The rise of GenAI and embodied AI agents provides a great opportunity to support researchers with this task. Yet, it is unclear how participants perceive interviews conducted by an embodied AI (i.e., an avatar), what level of embodiment is suited for this, and how much embodiment affects participants' engagement, interest, and openness to share information.

Goal of Thesis

The goal of this Bachelor/Master thesis (level will determine the scope) is to develop design knowledge about how embodied AI-Agents can improve participants' engagement, interest, and openness to share information.

1. Review of the related literature of existing systems in conducting interviews through AI (e.g., interview chatbots, social cues) and avatars.
2. Formulate hypotheses based on the related literature, e.g., users are more open to share when its appearance is robot-like, animal-like, object or human-like....
3. Implement avatar designs based on the literature.
4. Conduct an evaluation to assess how the designed embodied AI-Agents affect participants' engagement, interest, and openness to share information.

Requirements

• Experience with LLMs and prompting
• Strong interest in embodied AI agents in Virtual Reality
• You have read the following three papers: [1] [2] [3]