Most people still think of NotebookLM as a note-taking helper. A place to dump PDFs and get a quick summary. That mental model is now badly out of date, and the gap is worth your attention.
I just watched a walkthrough from Tina Huang, an ex-Meta data scientist turned creator, and my jaw was on the floor by the end. In her last NotebookLM video, over 10 months ago, the tool mostly summarized sources and made podcasts. In this new one, the creator shows it building docs, sheets, slides, infographics, videos, and even writing and running its own code. As she puts it, Google basically breathed general intelligence into a research tool.
Let me break down what she found, because the shift here is bigger than a feature list.
The old way vs the new way
Old NotebookLM: you upload your sources, it answers questions grounded in them. Useful, but passive.
New NotebookLM, based on what the author demonstrated:
- 🧠 It helps you find and import high-quality sources using the Gemini model, instead of you hunting them down manually.
- 📊 It writes and executes code in its own sandbox, so it can actually analyze data, not just talk about it.
- 🎬 It turns findings into slide decks, videos (anime style included), infographics, quizzes, flashcards, and reports.
- 🔗 It syncs with the Gemini web app and even has an MCP server for automations.
The difference is passive lookup versus an active research partner that produces finished work.
The business-idea demo that got me
Here is where the original poster made it concrete. She opened a featured notebook called The World Ahead 2026 by The Economist. These featured notebooks matter because the sources are pre-vetted by real experts, so the answers you get are trustworthy by default.
Then she gave it a sharp prompt. Something like: help me find market gaps to start a bootstrapped, automatable, solo B2C business that nets 20k a month within a year, and I can code and build with AI.
The notebook came back with four specific gaps drawn from 2026 trends:
- An AI daily companion for people on oral GLP-1 weight-loss pills, coaching them through side effects and muscle loss.
- A privacy-first log for the “California sober” and microdosing crowd, since wine consumption is in steep decline.
- A no-code tool for creators to generate vertical microdramas from text.
- A sandbox that teaches Gen Z graduates to build and deploy AI agents, since entry-level hiring is contracting hard.
What struck me is that every idea was tied back to a real trend and a real source. This was not the usual generic AI brainstorm.
From idea to full research report
The creator did not stop at a list. She showed a repeatable loop worth copying:
- She asked NotebookLM to find sources on evaluating micro-SaaS and solopreneur ideas.
- Then she pulled in deeper, industry-specific sources for each of the four ideas.
- Then she had it score each idea for viability using those bootstrapping and validation frameworks, saving each as a separate output.
One honest catch she flagged: you cannot clone a featured notebook the way you clone a Google Doc. So you have to copy the useful text into your own notebook by hand. A small friction, but real.
From those saved reports, the tool spun out a comparison slide deck, an anime-style video overview walking through a founder’s 30-day validation plan, and two infographics. The verdict it landed on was the GLP-1 companion as the top pick, with the “sober” wellness app as a strong second.
The part that was flat-out impossible before
Then the author switched to investing, and this is where the code execution shines.
She uploaded a snapshot of her real portfolio, exported The Economist’s global forecast table to a Google Sheet, and fed that back in as a source. Then she asked NotebookLM to find imbalances and opportunities by cross-referencing her holdings against 2026 predictions.
You could literally watch it run code to do the analysis. It surfaced things like a European defense spending surge, an India decoupling hedge as manufacturing leaves China, and a clean-tech angle. She was clear this is not financial advice, and that your results depend entirely on your own portfolio. But the workflow is the point: your private data plus vetted forecasts plus live analysis, all in one place.
How to try this yourself
If you want to copy the approach the expert showed:
- Start with a featured notebook so your sources are already high quality.
- Give it a detailed prompt with real constraints (budget, timeline, skills, target outcome).
- Ask it to find and import extra sources to go deeper on each angle.
- Have it evaluate and save each result as its own output.
- Convert the winners into decks, videos, or infographics to actually use them.
- For creative what-ifs that go beyond your sources, jump to the Gemini web app, since NotebookLM stays strictly grounded in its sources.
That last tip is a genuinely useful mental model. Use NotebookLM when you want answers locked to trusted sources. Switch to a broader chatbot when you want it to think freely with you.
The creator packed even more in, including a free video companion with these business ideas and a quiz at the end to help the concepts stick. If any of this sparked something, watch her full walkthrough for the exact clicks and prompts. It is worth the time.