01What Is the NotebookLM Update and How It Works

The new NotebookLM update adds customizable flashcards you can fix and share. In this episode, Julian Goldie explains how this small change makes a big difference for studying and working with AI.

Previously, NotebookLM created flashcards for you, but you couldn't correct them if they were wrong. Now you can change the question, fix the answer, add brand new cards, and share your deck with your team. The AI does the boring part, and you do the smart part.

02Main Features of the New NotebookLM Flashcards

The update brings several useful features to NotebookLM flashcards:

  • Fix cards: You can edit the question and answer on any card.
  • Add cards: You can create brand new cards yourself.
  • Share decks: You can share your flashcard set with your team.
  • Customize generation: Before generating, you choose easy, medium, or hard, set how many cards you want, and add a prompt.

This turns flashcards from a one-way tool into a collaborative workspace.

03How to Set Up the New NotebookLM Flashcards

Getting started with the updated Flashcards in NotebookLM is quick and requires no special skills. Julian shows these steps:

  1. Open NotebookLM and go to the Studio panel.
  2. Click on Flashcards.
  3. Click the pencil icon to open settings.
  4. Choose the difficulty level: easy, medium, or hard.
  5. Set the number of cards you want.
  6. Add a prompt to guide the AI.
  7. Generate your deck and review it.
  8. Fix any cards that need changes.

04How to Use the Flashcards and the Agent OS

Once you have your deck, you can use it for studying, training, or sharing with your team. But Julian shows a bigger use: he built NotebookLM into his agent OS, a dashboard with a team of AI agents.

Inside the agent OS, a research agent uses NotebookLM to read your docs and turn them into flashcards and key points. Then an SEO agent finds topics people search for. A video agent turns the best topic into a script using facts from NotebookLM.

Finally, a content judge checks the work and fixes weak spots.

Julian demonstrates this by growing his AI community: he drops notes into NotebookLM, the research agent learns it, the SEO agent finds topics, the video agent writes a script, and the content judge verifies it—all in one dashboard.

05Results and Comparisons From the Episode

Julian compares the old NotebookLM to the new one. The old version only made flashcards and you couldn't fix errors. The new version lets you edit, add, and share—so the AI and you work together.

He also shows the agent OS result: a full blog post and a full video for his AI community, made without babysitting. He says this used to take a whole day, but now it runs while he grabs a coffee.

The agent OS keeps all agents on one screen, and they share one memory. You set each agent's job once, and they go. No code, no tech headaches, no 50 open tabs.

06Tips, Limits, and Mistakes to Avoid

Julian gives several tips for using NotebookLM and the agent OS effectively:

  • Test the free flashcards first: Play with them to see how good AI has gotten.
  • Set clear jobs: Tell each agent exactly what to do, and they will work automatically.
  • Use the content judge: Always have a check agent catch mistakes before anything goes out.
  • Avoid overwhelm: Don't open 50 tools—put everything in one system.

A limit: the agent OS is a separate product, not a free tool. The flashcards are free, but the full agent OS comes with the paid AI Profit Boardroom.

07Who the NotebookLM Update Is For and Why It Matters

This update matters for anyone who studies, trains, or works with information. Students can make their own study cards. Teams can share decks to learn together.

Businesses can plug NotebookLM into a larger AI workflow.

Julian's agent OS shows how one tool can become part of a whole team. You can manage research, SEO, video, and quality checks from one screen. That saves hours and helps you get more customers with AI automation.

Most people do not know these tools exist yet, so acting now gives you an advantage.

08Other AI Agents You Can Build With NotebookLM

Julian also shows five useful AI agents you can build, all working with NotebookLM:

  • Research agent: Feeds it your docs; it makes flashcards and pulls key facts.
  • Content agent: Turns a topic into a full script you can press go on.
  • Lead agent: Finds new customers while you sleep and lines them up.
  • Helper agent: Answers the same questions over and over so you don't have to.
  • Check it agent: Reviews the work and tells you what's wrong.

These agents work together like a mini factory—NotebookLM reads, one finds ideas, one builds, one checks.