Colleagues in the History Faculty turned to AI and the assistance of the AI Competency Centre to try to address the overwhelming nature of academic research for undergraduate students who often struggle with where to begin when approaching complex historical topics.
Supported by the AI Teaching and Learning Exploratory Fund, the ‘Brainard the Fox’ project aimed to create a functional AI-powered research companion that could help undergraduate students navigate the overwhelming and complex scope of early medieval British history. The tool created was able to achieve this by suggesting and consolidating reading lists without providing students with complete essay solutions, giving them clearer pathways to engage critically with relevant texts.
This project was a technical collaboration between the History Faculty, the AI Competency Centre, and the Digital Innovation Team. From the AI and ML Competency Centre our Senior Research Software Engineers, Alok Sahu and Haseeb Ahmad, and Lead Business Technologist, Dominik Lukes, assisted in building the backend tool necessary to import and process documents to be collated into lists relevant to a student’s requirements. They did this through importing texts from Zotero and preparing them for bulk processing in Google Gemini for summary creation.
Following deployment of the prototype to a mix of staff and student evaluators, feedback suggested that the tool was highly valued in its provision of contextual overviews, its structured guidance features, and its ability to make approaching complex ideas less daunting. There was suggestion for improvement with the limitations of the narrow training dataset which led to repetitive and occasionally outdated recommendations, as well as evidence of a need to create further guardrails against the tool generating full-blown essay outlines for students.
After this 12 month project, ‘Brainard the Fox’ demonstrated some key learnings for the integration of AI in education. Primarily, this was around the importance of addressing pedagogical and ethical concerns in the department when integrating these sorts of tools - there is a critical need for clear usage guidelines and ongoing discussions about academic integrity in humanities when using AI. The project team concluded that AI tools work best when they intend to enhance rather than replace traditional academic skills.
Support for Neurodiverse Humanities Students
Project Case Study