Knowledge-Based Design Requirements for Generative Social Robots in Higher Education

TL;DR

Generative social robots (GSRs) powered by large language models enable adaptive tutoring but raise concerns about hallucinations and privacy. This paper argues that guiding GSR behavior requires specifying the robot's knowledge prerequisites, not only desired outcomes, and proposes a knowledge-based design framework categorized into self-, user-, and context-knowledge. Through twelve semi-structured interviews with students and lecturers, the authors identify twelve design requirements and explore how these knowledge types interact to produce responsible, effective tutoring in higher education. The findings offer a structured foundation for designing and evaluating GSRs in academic settings, emphasizing grounding in course content, personalization, privacy safeguards, and optional sensing, thereby balancing pedagogical impact with ethical considerations.

Abstract

Generative social robots (GSRs) powered by large language models enable adaptive, conversational tutoring but also introduce risks such as hallucina-tions, overreliance, and privacy violations. Existing frameworks for educa-tional technologies and responsible AI primarily define desired behaviors, yet they rarely specify the knowledge prerequisites that enable generative systems to express these behaviors reliably. To address this gap, we adopt a knowledge-based design perspective and investigate what information tutor-ing-oriented GSRs require to function responsibly and effectively in higher education. Based on twelve semi-structured interviews with university stu-dents and lecturers, we identify twelve design requirements across three knowledge types: self-knowledge (assertive, conscientious and friendly per-sonality with customizable role), user-knowledge (personalized information about student learning goals, learning progress, motivation type, emotional state and background), and context-knowledge (learning materials, educa-tional strategies, course-related information, and physical learning environ-ment). By identifying these knowledge requirements, this work provides a structured foundation for the design of tutoring GSRs and future evaluations, aligning generative system capabilities with pedagogical and ethical expecta-tions.

Figures (2)

• Figure 1: Knowledge Types Enabling Effective and Responsible GSRs
• Figure 2: Knowledge Types Enabling Effective and Responsible GSRs