Tutor CoPilot
Tutor CoPilot is an AI system designed to provide expert-level guidance to tutors during live teaching sessions, particularly in K-12 mathematics. By offering real-time teaching suggestions, it helps tutors develop and apply proven pedagogical strategies like guided questioning and conceptual scaffolding, rather than defaulting to simply providing answers.
The system aims to address a critical challenge in education: while expert guidance is crucial for developing effective teaching skills, access to such expertise is often limited and expensive. This particularly affects underserved communities, where students are most likely to work with inexperienced educators.
Tutor CoPilot represents a new approach to scaling educational expertise through Human-AI collaboration. At $20 per tutor annually, it offers an accessible way to improve teaching quality across diverse educational settings. While the system has shown promising results in improving student mastery of mathematical concepts, particularly with less experienced tutors, ongoing development continues to address challenges like grade-level appropriateness of AI suggestions.
Media Coverage
-
What Happens When an AI Assistant Helps the Tutor, Instead of the Student
AI-powered tutoring assistant increased human tutors’ capacity to help students through math problems and improved students’ performance in math -
This AI system makes human tutors better at teaching children math
The tool, called Tutor CoPilot, demonstrates how AI could enhance, rather than replace, educators’ work. -
How AI can improve tutor effectiveness
A Stanford University study of its AI tutor assistance tool revealed improved student performance and increased tutor capacity to support learning.
Related Publications
- Wang, R., Wirawarn, P., Khattab, O. ., Goodman, N., & Demszky, D. (2024). Backtracing: Retrieving the Cause of the Query. Findings of the Association for Computational Linguistics: EACL. https://doi.org/doi.org/10.48550/arXiv.2403.03956
- Wang, R., Ribeiro, A., Robinson, C., Loeb, S. ., & Demszky, D. (2024). Tutor CoPilot: A Human-AI Approach for Scaling Real-Time Expertise. arXiv preprint arXiv:2410.03017. https://arxiv.org/abs/2410.03017
- Wang, R., & Demszky, D. (2024). An Open-Source Library for Education Conversation Data. NAACL System Demonstrations. https://arxiv.org/pdf/2402.05111
- Wang, R., & Demszky , D. (2024). An Open-Source Library for Education Conversation Data. NAACL System Demonstrations. https://arxiv.org/pdf/2402.05111
- Demszky, D., Wang, R., Geraghty, S., & Yu, C. (2024). Does Feedback on Talk Time Increase Student Engagement? Evidence from a Randomized Controlled Trial on a Math Tutoring Platform. The 14th Learning Analytics and Knowledge Conference (LAK). https://spaces-cdn.owlstown.com/blobs/ttf4t8cpptbt692mxzao4p3d50af
- Wang, R., Wirawarn, P., Lam, K., Khattab, O., & Demszky, D. (2024). Problem-Oriented Segmentation and Retrieval: Case Study on Tutoring Conversations. Findings of the Association for Computational Linguistics: EMNLP.