<b>Experimentation and Potential for Generative AI Tools in Gateway Mathematics</b><b></b>

With the aid of tools like computer algebra systems, routine mathematical calculations and symbolic manipulations can be quickly and accurately performed. In the advent of rapid growth of Generative AI tools, we examine how ChatGPT performs in solving a variety of mathematical problem-types and methods of representations in gateway mathematics courses. Our survey on the current use of ChatGPT among learners and experimentation with this generative AI tool in a learning environment, compels us as educators to examine how ChatGPT incorporated in teaching and learning, can be used to enhance acquisition of factual, procedural and conceptual knowledge in gateway mathematics. Applying ChatGPT in mathematical discourse elicits some potential issues with both discipline-specific and general instructional practices while enhancing benefits associated with elements of High Impact Practices (HIPs). We focus on discipline-specific instructional practices that are characterized by problem-solving, problem-posing, open-ended questions with multiple solutions, modeling project activities, and technology integration. When applied to a class learning environment, the question of the quality of the ChatGPT output and what can be offloaded to ChatGPT becomes important to both learners and educators. Our analysis of the students’ survey and the experimentation with ChatGPT shows that using mathematical problem-based tasks, with instructor-curated content and learners equipped with refined prompts enriches the quality of gateway mathematics education in AI environments.