New AI Math Tutor to Help Students Problem Solve

Some educators may be hesitant to introduce artificial intelligence (AI) into their math classrooms, and for good reason. In some cases, AI can minimize human interaction, and give a solution instead of supporting the learning process. In others, AI can be an expensive and time consuming tool to integrate into your teaching routine. But what if there was a free, accessible AI tool that can provide tutoring support for students?

There are no known AI tools that currently support learning in the classroom and have been rigorously researched. As a leader in education technology, ASSISTments has been developing and conducting learning science research on ed-tech products and features since 2003. We are now poised to not only build a useful AI intervention but also to research its effectiveness.   

Multi-Million Dollar Grant for AI Math Research

Under a cost-sharing agreement, the DOE’s Institute of Education Sciences will contribute $3,749,600, or 90.89 percent of the funding, to the $4.1 million project, and WPI will contribute $375,815, or the remaining 9.11 percent. The funding will be used to develop a groundbreaking conversational artificial intelligence (AI) tutor named CAIT (pronounced as "Kate"), that will be integrated into the ASSISTments platform. This innovative program aims to support middle school students who struggle with math by providing them with personalized, interactive tutoring experiences. Worcester Polytechnic Institute will contribute more than $375,000, or the remaining 10 percent.

‍

How will CAIT work?

Through natural language processing, students will be able to speak or write questions for CAIT, and CAIT will respond with personalized replies. By monitoring students’ work, CAIT will be able to identify challenges each student is encountering and provide encouragement and additional problems for students to solve. This personalized approach is designed to foster independent learning and improve math proficiency among students who may not have access to traditional after-school or private tutoring.

‍

The project team at WPI is dedicated to ensuring that CAIT is not only technologically advanced but also pedagogically sound, integrating best practices in math education to maximize learning outcomes. WPI will create an in-house Large Language Model (LLM) to power CAIT, allowing it to be fully integrated within ASSISTments Teacher, which is freely available for teachers and students. This ensures that students have equal opportunities to access CAIT as a tutoring solution.

‍

“We are excited to work with WPI to implement generative AI into ASSISTments in such a meaningful and safe way. Positive experiences with math can boost students’ self-esteem and motivate them to keep going. This transformative solution will enable students to learn their homework instead of just ‘doing’ the problems.” - Cristina Heffernan, Executive Director of ASSISTments

‍

We believe that CAIT has the potential to revolutionize how students learn math, particularly those who need additional support. We look forward to the progress and impact of this exciting project and remain committed to advancing educational opportunities through innovative technologies.

Subscribe to our newsletter to receive updates on this groundbreaking AI tutoring program as it progresses!