In the rapidly evolving landscape of higher education, the integration of AI tools like ChatGPT into classroom practices has raised critical pedagogical questions. This study examines how undergraduate students experience ChatGPT as a peer-like presence within a Professional English course. Drawing on focus group interview with four students who interacted with ChatGPT during course activities, the research employs a reflexive thematic analysis to investigate how learners construct meaning around this AI-supported experience. Findings indicate that students frequently interpreted ChatGPT’s responses in relational terms, describing it as a supportive peer, group member, or teacher-like presence. Its immediate, non-judgmental feedback was seen to reduce anxiety, encourage participation, and scaffold experimentation with professional English. At the same time, role fluidity created moments of uncertainty and raised concerns about overreliance, diminished ownership, and “illusion of mastery.” Engagement was fostered through performative and collaborative tasks such as role-play and debates, which positioned ChatGPT as a co-actor rather than a mere tool. Confidence gains were strongest in structured tasks like vocabulary building and translation, while trust weakened in interpretive or research-oriented domains, underscoring the importance of pedagogical framing. These findings suggest that generative AI can provide pedagogical value beyond functionality, supporting relational meaning-making, scaffolding student confidence, and sustaining engagement when embedded in guided, collaborative practices. The study highlights the need for critical design of AI-integrated learning environments that balance innovation with reflective awareness of agency, trust, and ethical use.
ChatGPT student engagement relational interaction confidence higher education teaching and learning.
| Primary Language | English |
|---|---|
| Subjects | Instructional Technologies |
| Journal Section | Research Article |
| Authors | |
| Submission Date | September 10, 2025 |
| Acceptance Date | December 21, 2025 |
| Publication Date | December 30, 2025 |
| Published in Issue | Year 2025 Volume: 8 Issue: 4 - ICETOL 2025 Special Issue |