ArgueMate: An AI Debating Agent for Promoting Learning Through Argumentation
An AI-powered debate partner to enhance critical thinking with scalable and personalized learning experiences.
Duration: January 2025 – December 2025
Status: Ongoing
Educational Level: Tertiary Level
Topic: Artificial Intelligence AI, Digital Tools
Keywords: Arguing to Learn, Artificial Intelligence, Pedagogical Conversational Agent, Collaborative Argumentation
Initial Situation
Learning by arguing is a well-established pedagogical approach that enhances critical thinking and deep engagement with course material. Despite its proven benefits, scaling this method in classrooms poses challenges for educators, particularly in ensuring quality debates among students. Recent advancements in Artificial Intelligence, particularly Large Language Models (LLMs) like GPT-5, offer promising solutions through the development of Pedagogical Conversational Agents (PCA). In this project, we aim to investigate ArgueMate, an AI-powered debate partner designed to elicit argumentative discourse between students and agents. Unlike traditional peer debates, ArgueMate brings unique advantages of providing scalable, personalized arguing experiences by dynamically adapting its stance and argumentative style, aligning with pedagogical objectives of the educational practitioners.
Objectives
In this project, we aim to explore the design and impact of an AI-driven argumentation agent, ArgueMate, within educational settings. Our research is centered around understanding how such an agent can be effectively designed to engage students in peer-debating exercises and what the subsequent effects on their learning might be
Method
To address the research questions outlined above, we will adopt a structured Human–Computer Interaction (HCI) methodology encompassing learner-centered design, iterative development, proof-of-concept evaluation in the laboratory, and proof-of-value evaluation in classroom settings.
Planned Translation
The project aims to deliver a research-informed and field-tested prototype of the ArgueMate tool that is practically applicable for both individual and collaborative learning in real-world classrooms. The tool is envisioned to be publicly available as an open-source, web-based application, allowing for easy integration into existing educational platforms and curricula. We aim to translate the insights gained from developing and evaluating ArgueMate into actionable design guidelines and best practices. These outcomes will inform educators and researchers on how to effectively integrate AI-based arguing agents in classrooms. Impact will be assessed through classroom implementation outcomes, educator and student feedback.
