Professor Ari Seligmann demonstrated NotebookLM as a digital support for structured reading and note-taking, drawing on the platform's original note-card logic. The notebook example was built from two articles on a single topic, with notes taken alongside the readings. The framing was pedagogical: the goal was not just to use AI to summarise, but to model particular reading practices, extracting key quotes, recording observations, and then using the notebook to think comparatively across sources and notes.
The challenge was to move beyond passive summarisation and increase engagement with materials by using NotebookLM to support active reading, organised note-taking, and comparative thinking across multiple sources. Rather than asking AI to produce a summary and accepting it, the intent was to model a richer learning approach, where the users's own thinking is built into the knowledge base alongside the original sources.
I recommend opening the reading on one side of the screen and the NotebookLM notes panel on the other. As students read, they can take notes using key quotes, record observations, and reflect on the text, learning by reading and annotating, not just by asking questions and receiving summaries.
A distinctive feature of this approach was the ability to turn individual sources on and off, limiting the notebook's search and chat to only the selected sources at any given time. This allowed for focused engagement with a single text before zooming out to compare across the full set of sources.
The "magic" of NotebookLM in this context was the ability to convert notes into sources. Once notes were converted, students could have a conversation with both the original articles and their accumulated thinking simultaneously, building a layered knowledge base that grew more useful the more they engaged with it. From there, a prompt like "based on what I found interesting from the notes, give me some suggestions of something else" could extend their reading outward into related territory.
Watch the video below to see the NotebookLM outputs and what they looked like
This exemplar requires medium effort to implement.
Recommended resources and training:
This use case suggests that educators can teach students to use NotebookLM as a structured reading companion rather than a shortcut. It supports annotation, comparative thinking, and, if notebooks are shared, collective asynchronous learning across a group. More broadly, it models AI as something that supports disciplinary reading practices rather than replacing them. The knowledge base grows with the student's thinking, making it a tool for building rich understanding. From lecture content and lecture notes, to assessment information and brainstorming notes, to research sources and evolving notes and beyond this use case has a range of potential pedagogical applications.
Here are some additional resources that you can browse to help you implement this assessment.
