AI-Supported Feedback for Second Language Mandarin Learners: Affordances and Limitations Across YCT, HSK, and HSKK in Multilingual International School Settings

The rapid advancement of artificial intelligence (AI) has opened new possibilities for second language assessment, particularly in the context of Chinese learning and Chinese proficiency certification. This paper examines the potential of AI-supported feedback tools in addressing the persistent challenge of timely and personalized feedback for second language Mandarin learners across the full spectrum of Chinese proficiency examinations-the Youth Chinese Test (YCT), the Hanyu Shuiping Kaoshi (HSK, Levels 1–5), and the HSK Speaking Test (HSKK)-in multilingual international school settings. Drawing on second language acquisition (SLA) theory, feedback models, and classroom observations from international school environments in Singapore and the Philippines, this conceptual review analyses the affordances and limitations of freely available AI tools, including ChatGPT, HelloChinese, Duolingo Chinese, Google AI Studio, and Speechling, which support vocabulary acquisition, tonal accuracy, character recognition, oral production, and communicative competence. The paper further contextualizes these tools within multilingual policy frameworks, with particular reference to Singapore's bilingual education model and IBDP international school settings in the Philippines. Findings suggest that while AI tools offer significant potential for real-time, adaptive feedback, their alignment with YCT, HSK, and HSKK assessment criteria and the cultural-pragmatic dimensions of Mandarin remains limited. Pedagogical implications for practitioner-researchers and teacher educators in multilingual international school settings are discussed.