Using AI to Provide Feedback on Oral Presentations: Multiple Perspectives

Abstract Book of the 11th International Conference on Teaching, Learning and Education

Year: 2026

[PDF]

Using AI to Provide Feedback on Oral Presentations: Multiple Perspectives

Adam Forrester

ABSTRACT:

This study investigates the use of an AI based feedback tool to provide supplementary feedback on oral presentations for post graduate students at a university in Hong Kong. While AI has been widely adopted for feedback on written work, its application to oral coursework at the tertiary level remains relatively underexplored. The study examines the use of an AI feedback tool for both practice and assessed oral presentations.
A mixed methods approach was adopted. Data were collected from approximately 150 students through surveys and focus group interviews exploring perceptions of AI generated feedback. In addition, AI generated feedback was analysed and compared with teacher provided feedback to examine similarities and differences in focus, accuracy, and pedagogical value. The AI tool generated feedback across multiple dimensions of oral performance, including delivery features (e.g. body language, intonation, pausing, stress, and pronunciation), language use (e.g. filler words, repetition, and conciseness), and discourse organisation and audience engagement (e.g. signposting, flow, and engagement strategies). Perspectives from 10 teachers were gathered through surveys and interviews, including both teachers who used the tool and those who chose not to adopt it. Findings indicate that students were generally positive about the AI generated feedback, particularly in relation to pronunciation and delivery features. However, concerns were raised regarding the quantity and prioritisation of feedback, a view shared by teachers. Comparative analysis showed substantial alignment between AI and teacher feedback in identifying surface level delivery and language issues, while differences emerged in feedback related to higher level discourse organisation and communicative effectiveness. The findings highlight both the potential and limitations of AI based feedback for oral presentations and factors influencing teacher adoption in higher education.

Keywords: Higher Education Assessment; Speaking Performance; Pronunciation And Delivery; Student Perceptions; Teacher Adoption