THE ROLE OF CHATGPT, GRAMMARLY, AND DUOLINGO IN ENGLISH LANGUAGE EDUCATION: A COMPARATIVE ANALYSIS OF ARTIFICIAL INTELLIGENCE–POWERED LEARNING TOOLS

Abstract

Artificial intelligence (AI) has become an influential force in contemporary education, particularly in English language teaching (ELT), by enabling personalized learning, automated feedback, and adaptive instructional support. Among the growing number of AI-powered educational applications, ChatGPT, Grammarly, and Duolingo have emerged as three of the most widely adopted platforms for language learning. Although each application serves different pedagogical purposes, limited research has systematically compared their educational functions within a unified analytical framework. This study addresses this gap by examining the roles of these three AI tools in developing learners' English language competencies, including grammar, vocabulary, writing, speaking, listening, and learner autonomy. Employing a qualitative comparative research design, the study synthesizes findings from recent scholarly literature, international educational reports, and empirical studies published between 2023 and 2026. The analysis is guided by the Technological Pedagogical Content Knowledge (TPACK) framework and the Technology Acceptance Model (TAM), providing insights into both pedagogical effectiveness and technology adoption. The findings indicate that ChatGPT offers substantial support for interactive learning, personalized explanations, academic writing, and communicative competence. Grammarly demonstrates exceptional effectiveness in improving writing quality through real-time grammatical feedback, vocabulary enhancement, and stylistic recommendations. Duolingo contributes significantly to vocabulary acquisition, pronunciation practice, learner motivation, and sustained engagement through adaptive gamification techniques. Despite these advantages, each tool exhibits pedagogical limitations when used independently, highlighting the importance of complementary integration rather than substitution of traditional teaching practices. The study proposes an integrated AI-assisted instructional model that combines the strengths of these three applications to maximize language learning outcomes. The findings provide valuable implications for teachers, curriculum developers, educational policymakers, and higher education institutions seeking to implement AI technologies effectively in English language education.