Generative AI Tools in EFL Teaching and Learning: A Review with a Focus on the Iranian Context

The rapid growth of Generative Artificial Intelligence (Gen AI) tools such as ChatGPT, Gemini, Claude, and DeepSeek had started a new phase in English as a Foreign Language (EFL) education, although current literature on their role remains fragmented, especially in the Iranian context. Using the Technology Acceptance Model (TAM; Davis, ۱۹۸۹), the Technological Pedagogical Content Knowledge (TPACK) framework (Mishra & Koehler, ۲۰۰۶), and Sociocultural Theory (Vygotsky, ۱۹۷۸), this narrative review synthesizes existing studies to examine the role and implications of generative artificial intelligence tools in EFL teaching and learning, with particular attention to the opportunities and challenges these tools present in Iran. Consistent with Sociocultural Theory, the synthesis reveals that Gen AI tools mediate personalized, learner-centered, and interactive practice, with the strongest evidence found for writing development, whereas gains in speaking and listening remain less conclusive due to the limited pragmatic and social depth of AI-mediated interaction. In line with TPACK, the role of teachers is shifting away from routine instructional tasks toward facilitating critical thinking and meaningful interaction, a shift that depends on deliberate pedagogical integration rather than on technological sophistication alone. Consistent with TAM, adoption in Iran is shaped by unstable infrastructure, unequal digital literacy, and the absence of institutional guidelines, despite teachers generally holding positive attitudes toward these tools. Overall, this review points out that investing in technological infrastructure, improving AI and digital literacy, and creating clear institutional policies are necessary for the effective and ethical use of generative artificial intelligence (Gen AI) in EFL teaching and learning.