Welcome to the AI Ready Blog, where we explore the evolving world of Artificial Intelligence and Generative AI in education, fostering dialogue, experimentation, and research to enhance teaching, learning, and collaboration across disciplines.
Note: Features may change with future updates.
Welcome back to our hands on series on Google Gemini. In previous posts, we’ve discussed chain prompting and interacting with documents. This week, we’ll demonstrate how AI can assist with research, specifically using Google’s NotebookLM, a tool that lets you use Gemini in a grounded way, taking advantage of the LLM but only interacting with the sources you specify.
NotebookLM offers a different approach compared to some other generative AI tools. Its core function is to act as a research assistant, and a key aspect of this is its “grounded approach.” It uses only the source materials you provide (up to 50 documents) to generate responses, which helps to limit inaccuracies.
The accompanying demo video walks through setting up and using NotebookLM. We’ll look at a resource on Jesuit Pedagogy as an example. The video covers creating a new notebook, uploading source materials, and exploring the features. We’ll look at generating summaries and keywords, asking research questions, and using the Studio tab for study guides and FAQs. The video also includes a demonstration of the podcast feature. You can follow along at notebooklm.google.com.
Activity: Research Assistant Setup (NotebookLM)
Go to: notebooklm.google.com
NotebookLM is distinct from other generative AI tools in that it is a research assistant tool and its key feature is its “grounded approach,” using only user-specified sources (up to 50) to generate responses, minimizing inaccuracies (“hallucinations”).
Example Notebook: Resource on Jesuit Pedagogy
Steps: Create a New Notebook
- Go to NotebookLM.google.com
- Click “New Notebook”
- Upload relevant source materials (up to 50 documents)
- A note on selecting sources to add: whatever your topic, you might be tempted to only use primary sources. However, well considered secondary sources are just as if not more important because they are the ones that will most impact how NotebookLM summarizes and connects the points in your primary sources. So, a well thought out short article might be just as valuable as a long book.
- Now you can ask questions; create a faq and briefing document; create a podcast; and ask that questions in interactive mode.
- Be sure to interate from here. Your notebook doesn’t have to be static.Consider what sources you can add to make it better (or remove if you don’t like what’s it’s saying).
- Have fun!
Explore Key Features
- Sources tab
- Add Sources
- Generate summaries and keywords
- Chat tab
- Ask a research question
- Save important responses as Notes
- Studio tab
- Generate a study guide
- Create an FAQ
- Try the podcast feature with your materials
Conclusion
As the demo showed, NotebookLM has several potential applications in higher education. It can be used for tasks like literature reviews, understanding complex topics, creating study aids, and collaborative learning. Consider how NotebookLM might fit into your research workflow. It could be used to synthesize information from multiple papers, explore different perspectives on a topic, or generate learning materials. For collaborative projects, it could facilitate sharing and analyzing information from various sources.
I suggest exploring NotebookLM to see how it might be useful for your own work. Experiment with different source materials and features. In our next post, we’ll be exploring how to use AI to prepare for meetings, tests, and other scenarios where we need to stay on our toes and could use a practice buddy.
As always, feel free to reach out at cafferkey@fordham.edu if you have any questions.
Hope that helps!
Sponsored by the Fordham Faculty AI Interest Group | facultyai@fordham.edu | www.fordham.edu/AI
AI Events sponsored by the Fordham Faculty AI Interest Group