Implement educational chatbots with human-centred design, explicit AI policies, training, curated knowledge bases and continuous monitoring¶

Strategy · All strategies
Evidence · no claims cited

Description¶

The study recommends a package of implementation practices for educational chatbots in higher education: human-centred design (e.g., opt-in launch, clearer links to official resources), explicit AI policies and labels for assessment, staff and student training including prompt literacy, subject-specific knowledge bases, and continuous monitoring for errors and edge cases with subsequent updates of system instructions and knowledge bases. Academics should make clear which assessments may use AI, and chatbots can be instructed not to answer assessment questions or turned off during assessment exercises.

Design Implications¶

Context¶

Requirements¶

• Institutional policies on AI use, deployment and academic integrity
• Trained academics who know how chatbots are constructed and how to create appropriate knowledge bases
• Knowledge bases kept separate from generative AI training sets to preserve intellectual property and respect copyright and database contracts

Constraints¶

• The pop-up design was the largest annoyance to users, so automatic pop-ups should be avoided in favour of opt-in launch
• Chatbots are a supplementary tool, not a replacement for individual student-teacher interaction

Target Learners¶

• Higher education students, including international and distance-education students

Target Learning Goals¶

• Responsible and effective use of AI chatbots for academic support
• Maintaining academic integrity and critical thinking while using AI-assisted technologies

Affordances¶

• Jordan Chatbot Unit Specific Gpt4O Agent

Related Strategies¶

• Involve teachers as co-designers of AI-based intelligent systems, with agency to train, correct, interpret, and override system recommendations
• Implement sustained educator development programs including needs assessment, coaching, communities of practice, and experiential learning
• Train teachers in prompt engineering, evaluating AI-generated responses, ethical and privacy considerations, and integrating AI outputs into pedagogical decision-making
• Use a non-generative AI chatbot with a pre-programmed, human-curated knowledge base to answer student questions with pre-approved responses
• Implement custom AI chatbots as teacher-supported learning assistants with teacher oversight, curated databases, and AI literacy development
• Use AI chatbots to assist human-led coaching with routine support between sessions while humans retain challenge and goal clarification

Examples¶

-

Key Sources¶

• Stephen Colbran, Meena Jha and Caitlin Schiavone. (2026). Understanding student perspectives on generative AI chatbots: a human-centred mixed-methods study in higher education. International Journal of Educational Technology in Higher Education. https://doi.org/10.1186/s41239-026-00605-w