53 Comments
User's avatar
Suhrab Khan's avatar

NotebookLM isn’t just a summarizer, it’s a synthesis engine. Research papers, YouTube transcripts, industry docs… all connected into insights you’d miss on your own. This guide nails why 2025 feels like the golden age of AI tools. Worth experimenting!

Expand full comment
Geert van der Veen's avatar

We know that other ai agents lie to us. Can we trust this one? Why?

Expand full comment
Michael Spencer's avatar

Given that this is one of Google's best consumer AI products, and that their models have improved a lot since it was first released. I don't come across hallucinations very often on this product do you?

Expand full comment
Geert van der Veen's avatar

I read a lot about the work of socalled ‘quants’ in economics. These were extremely educated phycisists or other guys with a beta background. They were one of the important reasons behind the Huge economical meltdown that happened one decade ago. Its not that the Quants were lazy - they work extremely hard. Its not that they were unintelligent - all their calculations were spot on. The problem was more fundamental. The Quants and all the Banksters based their ideas on fundamentally flawed assumptions.They created an extremely logical and false intellectual castle in the sky. All of this led mr Greenspan to announce - short time before the horrile crash - that humanity had solved the basic economical questions.

It was nonsense.

They all believed it.

We were almost totally fucked.

Could AI ‘s not be the new Quants? In a very clever way - all their calculations and reasoning seem so perfect - confuse and corrupt us - and lead us away from the truth and down a dark path.

Especially because the current models - although admittedly often impressive - have no real understanding or intelligence whatsoever

( no disrespect to betas. They gave us quantum physics and are one of the backbones of modern society. We need them. There is this thing though - economics is not a hard science. This is why it is wrong to overly depend on Betas. We dont trust a surgeon to do the work of a psychologist - although both are highly intelligent, this is a truism yes)

Expand full comment
Geert van der Veen's avatar

Thank you. I will

Expand full comment
Michael Spencer's avatar

Extremely thought provoking comment, I invite you to read my next piece coming out in about 15 minutes.

Expand full comment
DistractedTimes's avatar

Helpful. Thank you.

Expand full comment
Nick Woodford's avatar

Thanks for the detailed explanation, I’ll definitely be trying the this with YouTube videos.

I used NBLM to create a chatbot for our internal sales team trained on all our research, vertical, and seasonal decks. It’s worked amazingly in reducing the amount of questions coming to the marketing team and saved them time scanning through multiple decks for the specific info they need.

Details here for anyone interested: https://open.substack.com/pub/marketsmarterwithai/p/turn-your-marketing-materials-into

Expand full comment
Eliza's avatar

I love this app

Expand full comment
Felicia Betancourt's avatar

This is the most horribly written prose I have ever read. Because the topic is so compelling, I tried to ignore one red flag after another until I stumbled onto this stinker: "How Chinese labs are building models like Qwen and DeepSeek-R1-Lite are only going to continue in 2025." Can that even be considered an English sentence? What is the subject of the second "are"???

Perhaps the authors (two people created this mess!) are too lazy or cheap to get a Grammarly subscription. Well, all I do know is that I am too busy to read their AI slop.

Expand full comment
Daria Cupareanu's avatar

I just love NotebookLM, there are so many ways you can play with it and extract insights.

Just last night, I ran a little experiment: I loaded all my Substack posts and a PDF of all the comments I’ve received, then asked it questions to understand what topics spark the most engagement, what themes keep coming up, and where readers want more depth.

I got great insights about my readers & a lot of fresh ideas for future posts

Documented it all here: https://substack.com/@dariacupareanu/note/c-119991656

Expand full comment
Alex McFarland's avatar

Amazing use case!

Expand full comment
Michael Spencer's avatar

How to create mind maps with NotebookLM: https://aidisruptor.ai/p/how-to-create-mind-maps-with-notebooklm

Expand full comment
TR's avatar

Great info. I'd like to use NLM in my work, but can't give it or provide access to sensitive company information. Any suggestions on how to utilize this tool interally? We have access to GCP. Maybe develop a similar workflow within this environment?

Expand full comment
Valerio's avatar

Really good quality post for newbie here!

Expand full comment
Darren Muirheid's avatar

In the spirit of sharing new ideas on how this can be used... I have been posting the output of deep research queries from chatGPT into NotebookLM as a source to absorb the complexity that comes out of deep research questions in an interesting way. One ? was on the state of the art for using random forests in diagnosing diseases where I can ask for the most successful applications and ask for new ideas on how to make the outcomes even better. Interesting.

Expand full comment
Mman's avatar

This is a super duper article. Really enjoyed it. Thanks for all the info!!!

Expand full comment
Loic Le Meur's avatar

Nice one thanks Michael

Expand full comment
aritechdiva's avatar

Write what your soul is speaking and whoever has the same frequency will respond back. The Law of reciprocity.

Expand full comment
Flyingmum/PamC's avatar

Share your tips on how you use it

Expand full comment
Timbo's avatar

I had made a mobile app to do this for YouTube to help me get summaries and the next step is synthesize things. I was building for myself but hoped to monetize my time.

Seems like even niche opportunities are going to be overtaken by big tech in the space.

Expand full comment