ARTIFICIAL INTELLIGENCE-DRIVEN TESTING IN EFL/ESP CLASSROOMS
This article explores the application of Artificial Intelligence (AI)-driven tools, particularly ChatGPT, for creating vocabulary test tasks in EFL/ESP classrooms. The research aims to evaluate the quality of vocabulary test tasks generated by ChatGPT by applying established criteria, including rel...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Institute for Digitalisation of Education of the NAES of Ukraine
2025-04-01
|
| Series: | Інформаційні технології і засоби навчання |
| Subjects: | |
| Online Access: | https://journal.iitta.gov.ua/index.php/itlt/article/view/5957 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | This article explores the application of Artificial Intelligence (AI)-driven tools, particularly ChatGPT, for creating vocabulary test tasks in EFL/ESP classrooms. The research aims to evaluate the quality of vocabulary test tasks generated by ChatGPT by applying established criteria, including relevance, reliability, interactiveness, practicality, and impact. It investigates how ChatGPT-generated tasks meet these criteria and provides practical recommendations for educators to optimize the quality of AI-generated assessments. The authors indicate that criteria such as relevance, practicality, interactivity, and impact can be fully satisfied in ChatGPT-generated tests. However, the research identifies challenges with the reliability of AI-generated test tasks, primarily due to ambiguities in response choices.
The article emphasizes the pivotal role of human intervention in guiding and refining AI-generated outputs. Detailed and context-specific prompts crafted by educators are critical to maximizing the potential of ChatGPT while mitigating its limitations. To support EFL/ESP teachers, the study offers detailed recommendations for enhancing ChatGPT-generated test tasks, such as developing precise prompts, setting clear contexts, assigning specific roles to ChatGPT, and iteratively refining outputs. These strategies improve the reliability and effectiveness of AI-generated assessments and align them with pedagogical standards. The authors emphasise the importance of integrating human oversight with AI tools to maintain the validity and usefulness of language tests. This research contributes to the broader discourse on integrating AI in education by demonstrating how educators can leverage ChatGPT for test design while addressing its limitations. Future directions include evaluating the effectiveness of other types of AI-generated test tasks, exploring AI’s role in automated assessment and feedback, and examining the long-term impact of AI-driven assessments on teaching methodologies and students’ vocabulary acquisition in ESP contexts.
|
|---|---|
| ISSN: | 2076-8184 |