-
The growing use of generative artificial intelligence (AI) is changing the ways multimodal resources are created and used in English language teaching (ELT). In primary classrooms, where visual and multimodal materials play a central role in supporting children's language development, AI tools offer teachers new opportunities to generate customized learning resources. Despite these developments, little is known about how primary English teachers evaluate AI-generated multimodal materials in comparison with textbook resources or how these evaluations shape their pedagogical decisions and AI literacy. Drawing on social semiotic multimodal theory and AI literacy frameworks, this study explores primary school English teachers' perceptions of multimodal resources in printed textbooks and AI-generated materials. It examines teachers' evaluations of these resources in terms of pedagogical value, multimodal design, cultural representation, age appropriateness, trustworthiness, and ethical considerations. Using a mixed-methods design, the study combines questionnaire data with semi-structured interviews in which participants compare textbook-based and AI-generated materials designed to achieve similar language-learning objectives. The study is expected to show that teachers appreciate the flexibility and adaptability of AI-generated resources while continuing to value textbook materials for their curriculum alignment, pedagogical coherence, and editorial quality. At the same time, teachers are likely to express concerns about cultural representation, bias, copyright, transparency, and the reliability of AI-generated content. These findings are expected to demonstrate that teachers' evaluation of multimodal resources is shaped not only by pedagogical considerations but also by ethical judgement and their capacity to make informed decisions about AI use in the classroom. The paper argues that AI literacy in primary ELT should extend beyond technical competence to include critical multimodal awareness, ethical reasoning, and teacher agency in selecting, adapting, and evaluating AI-supported learning materials. By highlighting these dimensions, the study contributes to ongoing discussions of ethics, agency, and the assessment of AI-generated educational resources, offering implications for teacher education, textbook development, and the responsible integration of AI into primary English language teaching.
