Teacher-AI Collaboration to Support Assessment and Feedback: A Case Study in Norwegian Secondary Education

Recent advancements in Generative AI (GenAI) demonstrate strong potential for enhancing student evaluation and assessment practices through complementary and symmetric human-AI collaboration. This study examines the dispositions, perceived benefits, and challenges that teachers experience when they and GenAI complement one another in providing formative assessment and feedback. Partnering with a leading GenAI-powered EdTech platform, we conducted a case study in a Norwegian secondary school using Learny, a GenAI-supported feedback tool. Qualitative data were collected from teachers who employed Learny in their feedback practices. The findings indicate that teachers were generally curious and open to using GenAI, while emphasizing the need to maintain professional responsibility. They valued GenAI's capacity to save time, offer inspiration, and deliver timely feedback, yet raised concerns about feedback quality, lack of relational and contextual awareness, and the risk of student over-reliance on such feedback. The study contributes to the growing literature on hybrid intelligence in education by identifying conditions under which GenAI can effectively support, rather than replace, human educators. It underscores the centrality of teacher agency in AI-supported feedback and assessment practices and outlines key implications for pedagogy, future GenAI system design, and sustainable implementation.