In partnership with
👋 Welcome to AI Central,
OpenAI's new Dots agents keep working on your tasks between conversations, in the background, while you get on with something else.
The catch is that you have to set the job up well before you walk away: clear steps, and the right sources to work from. That's what this week was about.
Today’s bonus guide takes it a step further. It shows you how to build your own AI custom brain in Gemini Notebook: an assistant that answers only from the PDFs, videos and links you give it, so you always know where an answer came from.
Which part of this week did you actually try? Reply and let me know - I’m so curious!
Below, you'll find this week in 60 seconds, the guide and what's coming next week.
Big hug and love you ❤️
Kris & The AI Central Team
🎙️ What's one thing you've done this week?
We’d love to hear your story. How the bootcamp is going, what you wish we’d cover or something you’ve built and want to show the community.
📊 Which AI type are you?
Are you a top 30%, 10% or 1% AI user?
1000s of readers have taken our quiz and discovered where they stand
Readers have used their score to spot exactly where to focus next and which areas to focus
It’s free and takes 40s - what are you waiting for?
📚 Want even more resources?
Most AI content online is noise - too basic or too niche to help with the job you're actually paid to do.
If you’re looking for extra material, the AI Central Library is the place for you: 3,000 hours of AI research, filtered into one place - searchable by topic, use case, and author, so you find exactly what you need in minutes.
There's a preview video on our homepage - watch it, then start a 1-month trial for $4.99 (less than a coffee).
👇 Get ahead before next week's guides
📋 Build your AI Custom Brain
This guide shows you how to use Gemini Notebook (formerly NotebookLM) to learn new skills faster by building a research workspace around expert knowledge. It walks you through adding PDFs, YouTube videos and website links, asking questions based on your sources and turning what you learn into AI-generated podcasts, flashcards and reports.
It’s useful if you want to organize expert knowledge in one place and turn it into a personalized AI learning system instead of spending hours searching through scattered resources.
Did you like this guide? Let us know and we'll build more like it.
📚 This week's recap
Chapter 16 - 5 steps for jobs too big for one prompt
You clear 40 minutes for the board update, put the whole thing in one message, and get 900 fluent words that decide nothing. The fix isn't better wording. It's gather, structure, draft, critique, revise - and knowing exactly where to put the cuts.
Chapter 17 - Turn today's prompt into a template
Where's your best prompt from last week? Buried in a thread you'll never scroll back to. How to spot the reusable half, mark the slots, and catch it in the 30 seconds before you touch the answer.
Chapter 18 - 6 questions to ask any long document
Nobody gets caught by a bad summary. They get caught by a good one. Six questions to put to a document instead, and a ten-second check that tells you whether an answer came from the page or from the model.
Missed one? No problem. They're written to be read in order but any of them stands on its own. Pick the one that sounds most useful and read it with a coffee.
📆 Where we're headed
This week was about documents and the prompts that handle them. Next week goes after the numbers inside them.
Monday. Why AI gets numbers wrong, and how to spot it before a figure reaches a slide with your name on it.
Wednesday. More on checking its work, this time where it matters most.
Friday. Getting it to argue with you instead of agreeing with everything you say.
