Stanford Working Paper Links Generative AI Homework Outsourcing to 18% Drop in Chinese College Entrance Exam Scores

A Stanford working paper released in June 2026, "The Generative AI Learning Penalty: Evidence from Chinese Secondary Education," uses data from about 27,000 Chinese students in grades seven to 12 to examine how self-directed generative AI use affects cumulative learning. It reports that roughly 80 percent of students began using generative AI between 2023 and 2025, and about 50 percent fully outsourced homework after five months of use. By the end of June 2025, college entrance exam scores fell 18 percent and high school entrance exam scores fell 24 percent, with no corresponding learning benefit found.

This is a rare study using large-scale real-school data to track the relationship between generative AI and exam performance, pointing to possible learning loss when students use AI to complete homework rather than whether the tools are used. For parents, teachers and schools, the link between homework outsourcing and lower exam scores needs to be considered in AI use rules; for AI education product developers, designs that mainly improve homework completion efficiency may amplify this risk. The paper is a working paper and has not been peer-reviewed, so causal interpretation and scope still require independent verification. Parents and schools need to reassess boundaries and supervision for students using generative AI to complete homework at home. AI education products centered on homework completion may face stricter school procurement and usage restrictions.