Presentation Schedule
Surpass Chatbots and Human Teachers? Explore the Effectiveness of Conversational AI in Language Learning (105029)
Session Chair: Dimitra Anastasiou
Sunday, 12 July 2026 15:40
Session: Session 4
Room: UCL Torrington, B09 (Basement Floor)
Presentation Type:Oral Presentation
Conversational Artificial Intelligence (ConvAI) has evolved from early rule-based systems and simple chatbots (e.g., Siri) to complex learning tools enabled by large language models, opening up new possibilities for language education. ConvAI is a type of AI that enables computers to understand, process, and respond to human language in a natural and human-like way. The introduction of ConvAI to language learning is especially timely and meaningful, as it overcomes the limitations of traditional learning methods, such as limited practical scenarios, mechanical or unengaging feedback, and high human costs. ConvAI systems can simulate real-life conversations, provide immediate correction, and offer adaptive learning options as well as context-aware feedback. This study evaluated the effectiveness and limitations of ConvAI in language education from two ConvAI-enabled language learning tools: Loora and Talkpal. The analysis reveals that ConvAI-enabled learning tools can break temporal and spatial barriers for language learning, offer highly personalized and adaptive learning experiences, provide vivid, immediate, and granular feedback to language learners, democratize learning costs, and ultimately enhance students’ grammatical accuracy and overall English proficiency. Our study also highlights key challenges in ConvAI-enabled language learning tools, such as limited depth, lack of human interaction, difficulties in fully replicating cultural and contextual nuances, model hallucinations, algorithmic bias, and data privacy risks. These risks may further erode user trust and pose obstacles to the widespread adoption of ConvAI. Finally, we compare learning practices between human teachers and ConvAI (humanoid) teachers. The comparison results are meaningful for the further improvement of ConvAI in language learning.
Authors:
Cong Qi, The Hong Kong Polytechnic University, Hong Kong
Yitian Sun, The Hong Kong Polytechnic University, Hong Kong
Shimeng Wu, The Hong Kong Polytechnic University, Hong Kong
About the Presenter(s)
Dr Cong Qi is a University Assistant Professor/Lecturer at The Hong Kong Polytechnic University in Hong Kong
See this presentation on the full schedule – Sunday Schedule
Conference Comments & Feedback
Place a comment using your LinkedIn profileShare this Presentation
Comments
Powered by WP LinkPress