This paper explores the effects of instruction that promoted AI literacy in English Language and Literature and Translation courses at an English as a Medium of Instruction (EMI) institution in southern China following the launch of ChatGPT-3, which prompted a global debate on the role of AI in education. It explores how pedagogical frameworks and teacher-student relationships shaped student engagement with AI while considering the technologization of education and its impacts on second language writing instruction. The study involved two different groups: Year 1 students in an Introduction to Corpus Linguistics course, and Years 3 and 4 students in a Learning through Digital Narratives course. Following the AI literacy framework suggested by Tseng and Warschauer (2023), the courses incorporated a variety of activities intended to facilitate students’ reflection on the affordances and limitations of AI tools and AI-generated products. Findings show that while AI tools can enhance text quality, they can also produce results that can be either vague or authoritative, potentially erasing the students' voice or style. The paper concludes by highlighting how teachers and students can benefit from their increased awareness of the potential and limitations of AI as well as the integration of AI into the curriculum to promote AI literacy in EMI higher education.