Connections Thematic Articles: Artificial Intelligence in Mathematics Teacher Education

Using ChatGPT as a Thought Partner in Writing Relevant Proportional Reasoning Word Problems

Andrea Berryhill (Mississippi State Univ.), Lendon Chandler (Mississippi State Univ.), Liza Bondurant (Mississippi State Univ.), and Bima Sapkota (Univ. of Texas, Rio Grande)

In this article, we, two preservice mathematics teachers (PSTs) and two mathematics teacher educators (MTEs), explore the potential of using ChatGPT, a large language model (LLM), as a thought partner to create relevant proportional reasoning word problems. We detail our iterative process of refining our inputs and critically evaluating ChatGPT's outputs. Our takeaways from this journey offer valuable insights for educators seeking to leverage the capabilities of LLMs in their lesson-planning endeavors.

Mathematics Teachers Using Generative AI to Personalize Instruction to Students’ Interests

Theodora Beauchamp and Candace Walkington (Southern Methodist Univ.)

Generative AI refers to a computer’s capacity to complete activities that usually require human intelligence, such as making choices, resolving issues, and acquiring knowledge. This innovative technology can allow teachers to personalize tasks to individual students and specific elements of their context. In this study, twelve teachers used three generative AI tools from Magic School to create personalized learning opportunities for students. They were asked to focus on the authenticity of problems, efficiency, differentiation, knowledge of specific interests, issues of bias, and student motivation. Through discussion, the teachers shared their insights. The benefits of using Magic School were personalized word problems and creative and functional ideas for primary students. Constraints and limitations were Magic School’s ability to create content focused on culture and language. More research should be done on Magic School’s many tools to continue understanding its functionality for all k-12 teachers. 

Beginning the Conversation: What is Responsible AI Use in Mathematics Education?

Kayla Chandler (East Carolina Univ.)

While Artificial intelligence (AI), particularly generative AI tools, has seemingly become popular in education, reception of AI tools has varied broadly, even within mathematics education. This paper briefly overviews some ways AI might be used in mathematics education and how that should inform and drive our work as mathematics teacher educators. With the many potential uses of AI, we must begin a conversation about what constitutes responsible AI use and move towards developing guidelines for best practices of AI use in mathematics education.

Bridging AI and Mathematics Teacher Education: A Teacher Educator’s Journey

Jonathan K. Foster (Univ. of Albany)

This paper reflects my journey in a multidisciplinary team developing an artificial intelligence (AI) tool for mathematics teacher education. I highlight the potential of AI, particularly computer vision, in enhancing teacher noticing via a video-based analytics dashboard. Mathematics teacher educators have a critical role in guiding AI development. This paper is an invitation to other mathematics teacher educators to contribute their perspectives on AI design and implementation.

Crowdsourcing Knowledge: Implications for Using AI in Mathematics Methods Courses

Juan M. Gerardo, Regan Blankenship, Sasha Cohn, and Julia Renfro (Univ. of Cincinnati)

I share my experience as an active Reddit (a social media platform) user, a lurker in the ChatGPT sub-reddit forums, who found inspiration to modify a three-phase lesson plan assignment for my mathematics methods course. With an example provided by three former students, they generated, critiqued, and analyzed a lesson plan produced by ChatGPT. Implications and concerns for both mathematics teacher educators and researchers are offered.

In the Shadows of Burgeoning Colossi: The Whiteness of AI in Mathematics Teacher Education

Carlos Nicolas Gómez Marchant (Univ. of Texas at Austin) and  Hamilton L. Hardison (Texas State Univ.)

It is important for mathematics teacher educators (MTEs) to consider AI in our practices. Without caution, MTEs may also be following (and leading the teachers they support) along a dangerous path. Inspired by Milner (2007), the authors explore field-specific dangers seen, unseen, and unforeseen regarding AI for MTEs. We foreground illustrative, but not exhaustive, dangers of each type: First, the seen dangers of equity issues identified and the race neutral policy discourse by professional organizations; Second, the unseen dangers of AI as perpetuating a (white) status quo; Third, the unforeseen dangers within the legal landscapes of AI.

Leveraging the Potential of AI as a Partner in Teaching

Nirmala Naresh (Univ. of North Texas), Zuhal Yilmaz (North Carolina State Univ.), and Terrie Galanti (Univ. of North Florida)

In this paper, we discuss our integration of mathematical modeling tasks and AI in a course for prospective secondary teachers. Our analysis of their reflections elicited a variety of pedagogical perspectives on the use of AI in their future classrooms. Experiences with mathematical modeling, coupled with AI as a collaborative partner, challenged the initial unease that many prospective teachers expressed about using AI to build conceptual understanding. Implications for MTE practice in fostering these pedagogical perspectives are discussed.

Computer Vision Activities for the K–8 Mathematics Classroom Emphasizing Feature Recognition Using Teachable Machine

Terri L. Kurz, Kimberlee Swisher, and Susan Jayasuriya (Arizona State Univ.)

Computer vision is an innovative technology that uses artificial intelligence tools like machine learning to analyze and identify objects. One popular tool for implementing computer vision in the classroom is Google’s Teachable Machine. Teachable Machine can classify items across a wide range of mathematics topics and grade levels with increased accuracy based on the number of images uploaded for training. Focusing on mathematics standards, three different activities that can be used in the K–8 classroom are described. Teacher educators are encouraged to explore the potential benefits of incorporating this technology into their instructional practices to support mathematical learning. Recommendations for teacher education are provided.

Counterexamples to Demonstrate Artificial Intelligence Chatbot’s Lack of Knowledge in the Mathematics Education Classroom

Amanda Sawyer and Zareen Aga (James Madison Univ.)

Mathematics preservice teachers are overconfident in the abilities of Artificial Intelligence chatbots like ChatGPT. Thus, we share counterexamples that mathematics teacher educators can showcase to demonstrate ChatGPT’s inaccurate mathematics information and inappropriate student materials.