01What NotebookLM source attribution is and how it works

NotebookLM source attribution is a new feature in Google's NotebookLM that shows you the exact sources behind every answer it generates. In this episode, Julian Goldie explains how it fixes the old problem of not knowing where the AI got its information.

Before, when you asked NotebookLM to create a study guide or a briefing doc, it was a black box. You got an answer but had no idea which file it came from. Now, with source attribution, you can click on any part of the answer and see the exact source and the prompt it used to build it.

It's like showing you the recipe for every answer.

This makes it super easy to check facts yourself. Instead of digging through all your files by hand, you just look at the sources right next to the answer. Done in two seconds.

02Key features of the NotebookLM update

This update brings two main features that change how you use NotebookLM. First is source attribution. Every time NotebookLM creates something, it shows you the exact sources it pulled from and the prompt it followed.

You can see the whole story behind every answer.

The second feature is the iterate button. This lets you tweak an existing answer without starting from scratch. You hit iterate, change one thing, and run it again.

It uses the same sources and setup from your original recipe.

There are also some smaller benefits. You can save recipes that work well and reuse them like templates. You can also make different versions of the same document for different audiences using the same recipe.

All of this is free and fast.

03How to set up and start with NotebookLM source attribution

Getting started with NotebookLM source attribution is simple. Julian shows how you can begin in just a few steps. Here is how to set it up and start using it.

  1. Go to NotebookLM and create a new notebook.
  2. Upload your files. You can use PDFs, notes, docs, or anything you want the AI to read.
  3. Type a prompt like "Make me a study guide about AI automation for the AI Profit Boardroom."
  4. Wait a few seconds for NotebookLM to build the document.
  5. Click on the generated text to see the source attribution.
  6. Check which file each line came from and the prompt it used.

That's it. You are now using source attribution. You can verify every fact in seconds.

04How to use source attribution and iterate with examples

In this episode, Julian walks through a real example. He uploads files about AI automation and his community, the AI Profit Boardroom, and asks NotebookLM to make a study guide. With source attribution, he clicks on a line and sees it came from a specific PDF.

No more guessing.

Then he shows the iterate button. Suppose the briefing doc is too long or too nerdy. He hits iterate and types "Rewrite this for a total beginner.

Keep it under 300 words. Add a quick bullet summary at the top." NotebookLM uses the same recipe but makes a cleaner version.

You can even make multiple versions for different audiences. Make a detailed one for your team, a shorter one for new members, and a tiny one for email. Same sources, same facts, but shaped differently.

This is the power of repeatable recipes.

05Results and comparisons from the episode

Julian compares the old way with the new way. Before, fact-checking was a nightmare. You had to dig through all your files by hand, one by one.

Most people just didn't bother. Now, the sources sit right next to the answer.

He also demonstrates how iterate saves time. In the past, if a document came out wrong, you had to delete everything, write a new prompt, and run it again, hoping it would be better. Now you just hit iterate and tweak one thing.

The results are cleaner documents, less time wasted, and better teamwork. Everyone can use the same recipe and get the same quality. Julian also notes that this builds trust because people can verify facts themselves.

06Tips, limits, and mistakes to avoid with NotebookLM

Julian is honest about the limits. Source attribution does not mean every word is perfect. It doesn't promise the AI is always right. What it does is show you where everything came from so you can check it yourself.

For important stuff like research, legal things, or health things, always double-check against your real files. Don't turn your brain off and just trust the robot. The feature makes checking easy, but you still have to do it.

Here are some tips to get the most out of it:

  • Use the iterate button to fix mistakes instead of starting over.
  • Save recipes that work well as templates for your team.
  • Make different versions for different audiences using the same recipe.
  • Always verify important facts against your original files.

07Who NotebookLM source attribution is for and why it matters

This update is for anyone who uses NotebookLM for real work. If you create study guides, briefing docs, FAQs, or timelines from your files, source attribution makes it faster and more reliable.

It matters because it turns NotebookLM from a black box into a transparent tool. You can see exactly where answers come from, which builds trust. When you share documents with your team or community, they can verify facts themselves.

This is also great for scaling your business. You can save recipes and let your team reuse them. Everyone makes the same kind of document the same way, every time.

No one has to guess.

Julian also mentions his free community, AI Success Lab, with over 38,000 members and 100 AI use cases. But the main point is that this feature makes AI actually useful for real work.