When educators restrict an AI model to a defined collection of trusted inputs, they can introduce students to academic materials in a flexible and interactive way. The approach, known as document-grounded AI, limits the model to sources such as lecture slides, assessment briefs, reading lists, and selected academic literature. The goal is to support inclusive learning by giving students a tool to revisit content, clarify difficult concepts, and prepare for assessments at their own pace.
Imran Khan, a senior lecturer in marketing at Birmingham City University, tested this method in a Principles of Marketing module with about 500 first-year students. The challenge was supporting a large cohort from different backgrounds. Some students lacked confidence in navigating academic expectations, interpreting assessment requirements, and engaging independently with module content. "The interaction with document-grounded AI provided an additional layer of academic support while complementing, rather than replacing, teaching and self-directed learning," he said.
Khan shared five practical steps for integrating document-grounded AI into teaching. The biggest lesson, he said, is not about the AI tool itself but how thoughtfully it is integrated into educational design.
Start with identifying the sources
Khan used a two-layer approach. Students first worked with core materials selected by the teaching team, then identified and evaluated additional academic sources through recognised databases. They uploaded these sources into NotebookLM and used the AI to ask questions, explore concepts, and compare arguments. The AI showed which uploaded sources informed its responses, allowing students to trace information back to the original evidence. This sequence keeps students responsible for finding and evaluating evidence while using AI to deepen their engagement with it.
Use AI to build confidence, not dependence
Khan integrated AI into seminars through short activities lasting 15 to 20 minutes. Students worked independently and then in groups, using lecture materials alongside additional academic sources they had identified and uploaded. Rather than simply revisiting notes, students could question the material, explore concepts, and use their understanding to contribute to group discussions. Tutors monitored engagement throughout the 12-week module and provided formative feedback during weekly activities. This gave less confident students opportunities to develop their understanding before contributing to discussions.
Develop skills to engage with information in different formats
Khan first used AI to create 10- to 15-minute video overviews summarising key concepts from one-hour lectures. During seminar activities, he encouraged students to use multimodal AI features to convert longer academic articles into shorter formats, including audio overviews and mind maps. This helped students digest complex concepts in different ways while continuing to understand the underlying academic content.
Introduce new ways to prepare for assessment
Document-grounded AI created a useful bridge between students and tutors during assessment preparation. Students could organise their research into a single source list or separate lists for different sections of the assessment. They were also encouraged to take notes during lectures and add these to the AI as sources. During seminars, tutors gained greater visibility of students' progress, including how they were engaging with research and using AI responsibly. Tutors could then provide formative feedback based on this activity.
Widen access for different learners
Supporting students from diverse educational backgrounds meant equal access to academic support while recognising that students engage with learning in different ways. The document-grounded approach provided a consistent foundation of trusted academic support while offering flexibility. Students could interact with the AI in different languages or through formats that suited their needs, bringing down barriers when dealing with complex concepts.
Qualitative feedback indicated that students valued this approach, particularly for strengthening their understanding of complex concepts and preparing more effectively for assessment. Module outcomes were also encouraging, with the awarding gap reducing from 21.4 per cent to 18.2 per cent. Although these findings do not establish causality, they suggest that the approach may contribute to more inclusive learning as part of a broader pedagogical strategy. For educators looking to build similar skills, an AI Learning Path for Teachers offers practical training on integrating these tools into classroom settings.
Why this matters for educators
Document-grounded AI shifts the conversation from policing AI use to designing structured interactions with trusted sources. The approach does not replace teaching or independent research. It adds a layer of academic support that students can access on their own terms, in their own languages, and at their own pace. For instructors managing large, diverse cohorts, it offers a way to make formative feedback more visible and targeted without adding grading hours. The key is starting with a clear educational problem, then selecting sources and designing activities that keep students actively responsible for evaluating evidence.
Your membership also unlocks:
