Conference
Open access
Methodologies for Improving the Quality of AI Tutoring in K-12 Education
Tushar Udeshi
Anna Khazenzon
Kabir Khan
Nick Breen
R. J. Corwin
Chris DiGiano
Kodi Weatherholtz
Marek Zaluski
Aug 2026 · International Conference on Artificial Intelligence in Education · pp. 201-214 · 0 citations
· 26 references
Computer Science
Abstract
Many AI tutors leverage large language models (LLMs) today. Given that LLMs are opaque black boxes, robust evaluation and live experimentation to measure the impact of every change are essential. We pioneered AI-powered tutoring for K-12 with the launch of Khanmigo (Khan Academy, 2023). We describe the metrics we use to measure AI tutoring quality and student engagement as well as various experiments we have run. We highlight the changes that have moved our metrics, including models, prompting, personalization and agents.