Benefits and Potential Issues for Intelligent Tutoring Systems and Pedagogical Agents

Abstract

When intelligent tutoring systems (ITSs) were first introduced, they were considered efficient and effective tools that helped students learn. Their effectiveness is typically measured against human tutors, generally treated as the ideal format for instruction. After decades of development, state-of-the-art ITSs are already as good as human tutors, at least in some lab studies, and in some cases substantially better. Advances in ITS technology have now inspired some concerned parents to question the potential harm of using ITS, specifically when young students interact with anthropomorphic ITSs. In this chapter, we critically evaluate ITS, especially those that mimic real humans (such as conversation-based ITS), their potential role in helping students learn, and most importantly, potential harm to human learners. We propose that ITS researchers and developers must assume the responsibility to understand the effect of prolonged exposure to anthropomorphic agents and prescribe ITS usage accordingly.

Publication Title

The Frontlines of Artificial Intelligence Ethics: Human-Centric Perspectives on Technology’s Advance

Link to Full Text

Share

COinS