Evaluating Generative AI in Education: LLM as English Tutor
摘要
Generative Artificial Intelligence (GAI), powered by Large Language Models (LLMs), shows promise as an assistive technology for English language learning in Brazil. Facing student challenges and limited technological resources, GAI emerges as a tool for personalized and contextualized learning. This article evaluates an intelligent tutor’s implementation, exploring its content adaptation, teacher moderation, interactive corrections, and resource access capabilities. The study employs an experiment with students to assess the intelligent tutor’s learning support capacity, followed by a qualitative evaluation. The objective is to identify GAI’s limitations and potential in supporting English learning, seeking solutions for challenges faced by Brazilian students.