Teachers configure AI chatbots but intent often differs from outcome
Will It Teach as Intended? How Teachers Configure Educational AI Chatbots
Human-Computer Interaction
Summary
Teachers use AI chatbots to support their students, adjusting settings to match teaching goals. This study found that while teachers try to make chatbots behave in helpful and personalized ways, the chatbots often don’t perfectly follow these plans. The researchers observed that the chatbot’s responses matched teachers’ expectations best in how friendly or responsive they were, but less so in following detailed teaching rules or goals. This means tools need to improve to help teachers better control what the chatbot actually does.
What this means in practice
- •For educational software developers: Design chatbot authoring interfaces that better capture and test teacher intentions for educational use.
- •For customer support chatbot teams: Improve chatbot configuration tools to ensure customer service bots follow intended interaction rules and goals reliably.
Authors
Bahare Riahi, Deniz Ozturk, Alice Guth, Jiayu Li, Daksh Pratap Singh, Xiaoyi Tian, Jennifer Chiu, Nicholas Lytle, Tiffany Barnes, Veronica Catete
Abstract
Teachers are increasingly using generative AI to support instruction, yet it remains unclear how pedagogical intentions are translated into chatbot configurations and reflected in chatbot behavior. We studied a teacher-facing chatbot authoring tool in professional development workshops with 27 middle school teachers, analyzing focus-group interviews alongside configuration and interaction logs. Teachers envisioned chatbots as instructional scaffolds that could provide differentiated support, extend access to assistance, and preserve student thinking within teacher-defined boundaries. Configuration analysis showed that Purpose primarily captured instructional goals and content focus, whereas Rules more often specified pedagogical behavior, guardrails, and learner-specific adaptations. Log-based evaluation showed stronger alignment for responsiveness (88.9%) and persona (81.5%) than for rules (70.4%) and purpose (59.3%). These findings show that configurable controls alone do not ensure pedagogical fidelity and highlight the need for authoring tools that help teachers express, test, and refine intended chatbot behavior.