No items found.
No items found.
No items found.
No items found.
Endorsed
*This resource has been tested for appropriateness in the classroom and scrutinised for safeguarding and cybersecurity issues. However, please do carry out any due diligence processes required by your own institution before using or recommending it to others.
Experimental
*This is an example of a resource that is under development and may not have been fully vetted for security, safety or ethics.  Please carry out your own safeguarding and cybersecurity due diligence before using it with students or recommending it to others.

Leveraging AI for Strategic Learning: Designing Custom Chatbots

Secondary
Sixth Form
No items found.
Teaching & Inclusive Practices
Key Stage 3
Key Stage 4
Key Stage 5
No items found.
Case Study
No items found.
No items found.
Practitioners
Joe Cozens

Transformation & AI Lead

This case study explores the implementation of custom AI chatbots, developed by Microsoft Copilot, to create a safer and more strategic learning environment. The aim was to test for hallucinations and evaluate the effectiveness of closed-loop environments, enabling students and staff to have confidence in the reliability of AI-generated responses.

A suite of student-facing and teacher-facing AI chatbots were developed and integrated into the Microsoft Teams environment. These chatbots were designed to operate within closed-loop environments, sourcing information from approved documents, SharePoint sites, and specific websites.

 

For students, the chatbots included:

 

- AQA GCSE Biology Chatbot: Provided answers sourced from authorized materials for the AQA GCSE Biology curriculum.

- Biology QuizBot: Generated quiz-style questions and answers for students to practice Biology concepts.

- UCAT Exam Chatbot: Supported students preparing for the University Clinical Aptitude Test (UCAT) for entry into medical programs.

- UCAS Chatbot: Assisted students in finding information about university destinations and courses.

- Scholar Research Chatbot: Sourced answers from scientific journals to support student research inquiries.

 

For teachers, the chatbots included:

 

- Safeguarding Chatbot: Provided guidance on potential safeguarding scenarios by referencing internal safeguarding policies.

- Document Finder Chatbot: Retrieved information from the school's internal SharePoint site and external website, enabling efficient access to relevant resources.

 

This implementation allowed for controlled testing of AI-generated responses, ensuring factual accuracy, usefulness, and ethical considerations were addressed within the closed environments.

The scenario in this case study is genuine and based upon real events and data, however its narration has been crafted by AI to uphold a standardised and clear format for readers.

Key Learning

The custom chatbots proved effective in theoretical scenarios, demonstrating the potential for strategic use of AI in educational settings.

However, the full capacity of these chatbots was not extensively tested. The future adoption of custom chatbots in schools may depend on the availability of skilled IT staff for development, as well as the possibility of centralized, government-controlled or exam board-provided chatbots for widespread use across institutions.

Risks

Cost: Designing and maintaining custom chatbots would incur ongoing expenses for schools.

IT Staff Skill Capacity: Retaining skilled IT personnel capable of constructing and managing these chatbots could be a challenge due to potential staff turnover.

Trust and Safeguarding: Safeguarding matters should not be solely delegated to AI systems. Until AI chatbots are approved for use with minors, direct student access may need to be limited, with testing conducted on staff computers for development purposes.