AI-Powered English Learning Platforms: A User Experience Evaluation

Abstract

The rapid development of artificial intelligence (AI) has transformed English language learning through adaptive and personalized platforms such as Duolingo, ELSA Speak, and ChatGPT. While previous studies have focused mainly on learning outcomes and motivation, limited attention has been given to learners’ holistic user experience (UX). This study examines learners’ perceptions of usability, enjoyment, and accessibility in AI-powered English learning platforms and explores how contextual factors, including digital literacy, learning styles, and prior AI experience, influence user engagement. A mixed-methods design was employed. Quantitative data were collected using a User Experience Questionnaire (UEQ) from 77 Indonesian EFL university students with at least three months of experience using AI-based platforms. Qualitative data were obtained through semi-structured interviews with five purposively selected participants. Descriptive statistics and correlational analysis were applied to quantitative data, while thematic analysis was used for qualitative data. The findings indicate positive perceptions across all UX dimensions, with accessibility receiving the highest rating (83%), followed by usability (81%) and enjoyment (80%). Task efficiency (76%) and visual design (73%) emerged as areas requiring improvement. Digital literacy and prior AI experience significantly influenced perceived usability and overall satisfaction. Qualitative findings highlight the importance of immediate feedback, adaptive features, and flexible learning environments, while also revealing concerns about reduced human interaction. These results underline the need for user-cantered design to ensure sustainable and inclusive AI-supported language learning.