"Bespoke Bots": Diverse Instructor Needs For Customizing Generative AI Classroom Chatbots

Irene Hou, University of California San Diego

Philip J. Guo, University of California San Diego

Date

02/2026

AI Overview

Why it matters

A single classroom chatbot cannot fit all instructional contexts. Designing reusable, effective AI assistants requires understanding how instructors prioritize and decompose customization needs across varying course sizes, disciplines, and teaching styles.

Study at a glance

Question: How do university STEM instructors prioritize and adapt customization features for generative AI classroom chatbots?
Setting and data: 182 educational prompt resources; 10 university STEM instructors in North America and Europe, June to August 2025.
Method: Qualitative content analysis and semi-structured interviews with a card-sorting activity.
AI tools: ChatGPT, Claude, Gemini

Key findings

Instructors consistently prioritized grounding chatbots in course materials and pedagogical strategies, while viewing safety constraints and guardrails as a necessary baseline.
Instructors consistently de-prioritized chatbot persona and communicative tone, with several expressing concern that personable AI could harm student-to-student interaction.
Needs diverged widely regarding personalization, course management tasks, and user platforms, driven heavily by class size, course structure, and subject matter.

Summary drafted with AI from the full paper and may contain errors. See the full paper for details.

Instructors are increasingly experimenting with AI chatbots for classroom support. To investigate how instructors adapt chatbots to their own contexts, we first analyzed existing resources that provide prompts for educational purposes. We identified ten common categories of customization, such as persona, guardrails, and personalization. We then conducted interviews with ten university STEM instructors and asked them to card-sort the categories into priorities. We found that instructors consistently prioritized the ability to customize chatbot behavior to align with course materials and pedagogical strategies and de-prioritized customizing persona/tone. However, their prioritization of other categories varied significantly by course size, discipline, and teaching style, even across courses taught by the same individual, highlighting that no single design can meet all contexts. These findings suggest that modular AI chatbots may provide a promising path forward. We offer design implications for educational developers building the next generation of customizable classroom AI systems.