How to use Generative AI for reading and note taking

Introduction to AI for reading and note taking

AI can be a powerful tool for supporting your reading and note taking, however, it must not replace your own reading and note taking skills. Instead, use AI tools, when appropriate, to help you organise your thinking, generate questions, and keep track of key arguments. 

There are a range of AI tools that can be used to support reading and note taking. This includes Google Notebook LM, which all students and staff at the University of Sheffield have access to via Google Workspace. If you need further information on how to access it consult the IT Services web pages.

Use of Notebook LM may not be appropriate in every context. Often, it is essential to read, engage with and respond to the text yourself rather than using AI to assist you. Before using any AI tool, always check the guidance on AI use within your school, review assessment briefs, and speak with your tutors if you are uncertain. Also, remember never to copy and paste any AI output into your work. Doing so would leave you at significant risk of academic misconduct. 

How does Notebook LM differ from other AI tools?

Notebook LM and similar tools differ from other GenAI tools in a number of ways. 

Opportunities and challenges of using AI for reading and note taking

Opportunities 

• Tools such as Notebook LM are designed to work directly with the texts you upload and  will usually ground its responses in the material at hand, reducing the likelihood of hallucinations.
• These tools are particularly well-suited for helping with reading, and can clarify complex arguments, track recurring themes, and help you compare different sections of a book or article.
• AI tools can provide notes and summaries of texts in a range of different formats. For example, Notebook LM can be used to generate audio, including in the form of a Podcast. This might be useful if you learn well from listening. However, it is important not to rely on a computer-generated podcast as an alternative to close engagement with the original text.
• AI tools can be a helpful revision tool that can allow you to generate flash cards, mind maps or quizzes based on texts or notes that you have provided as source material. 

Challenges

• AI tools do not provide a substitute for your own reading, critical thinking and analysis. Familiarising yourself with a text first, before introducing an AI tool, will allow you to get the most out of the tool as a research aid. Visit the Reading and understanding information hub to find out more.
• AI tools such as Notebook LM can still hallucinate. These tools generally work best when asked specific questions about the contents of the text you have uploaded. The further you move away from these interrogative questions, the more likely a hallucination may occur. Sometimes, a response may indicate that the AI tool is guessing, e.g. ‘this likely refers…’ Such wording places all the more importance on verifying through careful reading of the original, and making an informed judgment yourself.
• These tools work best for secondary sources that follow the conventions of academic writing, e.g. texts with clear introductions that pose questions, set out arguments, and are made up of paragraphs that develop a single main point. The more experimental or unconventional the writing, the less likely these AI tools are to generate helpful responses.
• You may disagree with the AI tools outputs, and that’s good! Like any other AI, tools such as Notebook LM do not work in the same way as our brains, and do not read a document in the manner of human intelligence. Using these tools as a fallible assistant that aids your comprehension, rather than relying on it to comprehend for you, is a sensible approach. 

Getting started with using AI for reading and note taking

The step by step advice below provides a walk through of how to use AI to enhance your approach to reading and note taking. As all students  at the University have access to Notebook LM, this tool is used in the examples provided below. 

1. Start with a minimum of a skim read of the text

   • What to Do: Before uploading a text into an AI tool such as Notebook LM, undertake at the very least a 15-minute skim read through the article or chapter. Look at the abstract, headings, subheadings, introductions, conclusions, and topic sentences. You may also read the text more carefully and/or write a short paragraph or set of bullet points capturing what you have taken away from the text and what further questions it has raised.
   • Goal: You are trying to form an initial sense of the author’s purpose, scope, and key arguments without the aid of the AI tool. Note down terms, issues, or events that seem important as well as questions or points of confusion. This stage is essential to get the most of these tools - and to avoid overreliance on the AI outputs.
   • Risk: missing out this step will leave you unfamiliar with the overall structure and tone of the text, making it harder to verify ideas or report on the text later on.
      

2. Upload a text to NotebookLM

   • What to do: Upload a text in PDF format or via a weblink to Notebook LM, which will generate a summary and suggest questions (providing the text is legally acquired). View the Generative Artificial Intelligence (genAI) and Copyright guidance to find out more.
   • Goal: uploading a text to NotebookLM will allow you to analyse it further by posing your own questions.
   • Risk: While convenient, the automatically-generated summary and questions can oversimplify, omit crucial issues, and above all discourage you from engaging actively with the text.  Consider ignoring them altogether at this stage. Relying on them is likely to hinder, not help, your comprehension.
      

3. Pose new questions to aid comprehension

   • What to do: Go back to your own initial read through  (and the guidance and topic related to individual seminar sessions, etc. that you are preparing for), and use that as the basis for formulating question prompts.
   • Goal: Questions of your own are much more likely to help your understanding than ones generated for you.
   • Risk: AI tools such as Notebook LM work best when prompted with thoughtful, interrogative questions about the text itself. Vague prompts produce vague answers. What, why, how, and where questions can be especially useful. 

4. Verify, read the original closely, and take careful notes

   • What to do: One of Notebook LM’s most useful functions is that it directs you to text related to the questions you’ve asked.  Most responses to your prompts will include numbers in grey circles. Click on these to locate relevant sections in the original source. You should now read those passages - and the surrounding text - carefully in the document  you have uploaded.  Note down any key points in your own words, record key quotations with page numbers and clear quotation marks, and jot down ideas for discussion / reflection.
   • Goal: this ensures that you check the accuracy of Notebook’s output, undertake a closer reading of the original text and produce your own usable notes.
   • Risk: Lifting notes directly from Notebook LM responses is quicker but comes with considerable risks; if you do so, your notes must clearly show where you are copying and pasting from the LLM. You are more likely to understand and remember information when putting arguments into your own words, and clear quoting and paraphrasing reduces the risk of academic misconduct. 

5. Juxtapose multiple texts

   • What to do: One of the most powerful features of Notebook LM is the ability to upload two or more texts at once.  Upload the relevant articles or chapters together, then ask Notebook LM to draw connections or contrasts. Both texts must be selected.
   • Goal: This allows you to see how different scholars frame similar issues, use evidence, or position themselves within a debate.
   • Risk: Use Notebook LM’s comparisons only as a starting point - always return to the texts themselves to verify and deepen your understanding and to ensure that your notes are accurate.  

Top Tips

• If you use Notebook LM as an aid to help you read and understand, rather than a replacement for reading, you will be well-placed to make the most of the tool. Employing it well should develop your information literacy and note-taking skills as well as your comprehension of texts.
• It is also important to review the University page on Generative AI and Academic Integrity before using any AI tool to help you in your studies. Use of AI, including Notebook LM, must be declared via your School's agreed process, such as the Acknowledge, Evidence, Describe coversheet.. NotebookLM chat history is not saved, you would have to create a note to retain a record of what the output said in response to each individual prompt, and keep a record of the prompts used.
• You should also read any guidance from your school and the assessment brief before using any AI tools. Some assessments may prohibit the use of AI in part or full.

With thanks to Dr  Andrew Heath, School of History, Philosophy and Digital Humanities at The University of Sheffield who developed guidance on using Notebook LM, much of which has been used to inform the development of this page. 

Next steps 

• How to use Generative AI for assessment
• How to use Generative AI for essay and report writing
• Using AI and GenAI to find information

Further Resources 

• Principles for Generative AI in learning and teaching