In March, we launched the inaugural cohort of The AI Second Brain (AISB), a new immersive education program years in the making.
It marked my return to teaching after three years of intensive research and experimentation with AI. I felt like an explorer returning from the wild frontier with a beacon of hope in hand – a new way of working with AI that was centered on augmenting humans, not replacing them.
I honestly wasn’t sure how it would go. I didn’t know if there was still an appetite for human-to-human education, as many were predicting its demise (and still are). I wasn’t sure if the approach I’d developed for myself would resonate as much with others. Most of all, I wondered whether I could effectively teach it to a group of people from a wide variety of backgrounds.
To my amazement, more than 350 people answered my call from over 70 countries – founders and business owners, software engineers & designers, consultants and strategists, executives and coaches, surgeons and lawyers, product managers and marketers.
They hailed from dozens of different fields – technology and software, healthcare and medicine, finance and banking, management consulting, education, law, marketing, real estate, construction, and government and the public sector.
We had professionals join from some of the world’s most important companies and organizations – Amazon, Google, Apple, Samsung, Morgan Stanley, EY, RBC, UnitedHealth Group, Dow, Blue Origin, SEAT, Match Group, Arista Networks, Colliers, and the U.S. Bureau of Ocean Energy Management.
We all spent 3 weeks together in a hands-on sprint designed to build their own AI Second Brain piece by piece, not just learn abstract concepts. After it was all said and done, graduates gave the program a Net Promoter Score of 70, rivaling the strongest brands in the world like Apple and Google.
But what really hit home for me the most was hearing about the alumni’s experience in their own words:
Elaine said that AISB “advanced my knowledge base and my ability to work with AI… at least 10-fold.”
Nick said “I figured it was probably just a collection of materials I could piece together from online sources on my own. I was wrong. This class is far more than the sum of its parts.”
Alison loved how our approach “keeps you, the human, firmly in control.”
Tim’s favorite part was “how to show up in my body and human experience when interacting with this powerful technology.”
But I think my favorite quote came from Maggie — “the sessions build on each other in a way that actually compounds… you have a system.”
That’s what I was always missing, going way back to the early days of my career when I first discovered GTD and realized that you could build a system around your work. That idea inspired my career searching for a similar system for digital content, knowledge, and learning.
And today, the need for a systematic approach to AI is more glaring than ever. Every feed is inundated with impressive tips and tricks and the latest feature that is supposed to “change everything,” but no one is offering a principles-based and holistic path through the storm of change.
It’s why I created The AI Second Brain in the first place, and why we’re running it again this fall, from Sep. 17 – Oct. 8. Our mission is to transform your relationship with AI, from seeing it as an uncertain, overwhelming, or even threatening force, to feeling it power your very own cognitive exoskeleton.
The 10 Updates We’re Making to The AI Second Brain
There’s been tremendous progress in AI over the last 5 months since we came together for the first cohort in April.
My team and I have spent the last few weeks in a deep dive, finding the most impactful and important developments to share with you as part of Cohort 2. Here’s what you can expect from it, whether you’re a returning student or a new one:
#1. Alumni Mentors leading dedicated Implementation Labs
The single biggest change we’re making is the introduction of Alumni Mentors as part of a wider teaching staff. We’ve run a competitive application process to identify the 8 graduates of the program who have the most to offer new students in terms of their knowledge, experience, and expertise, and are training them in the coaching and facilitation techniques we’ve developed over many years.
Each Alumni Mentor will lead their own Implementation Lab, an interactive space designed to support you through applying what you’ve learned in the program. They’ll offer examples, guidance, and ways of adapting the implementation steps based on their background and experience – on different tools like ChatGPT or Microsoft Copilot, for different use cases such as vibecoding or design, or for different life stages such as retirement or student life.
I’m tremendously excited to see how they translate the course material to different contexts and help people apply it to a wider range of use cases.
#2. Active attention practice
It’s always been important that you know how to direct and sustain your attention. But the game has changed, and it’s now time to go up a level: to notice what you like, and why you like it. When something in your field of perception attracts you, is it because it’s resonating with something already inside you, or because it’s hijacking your mental biases?
I’ll introduce you to a set of tools for “Active Attention Practice,” all of them designed to sharpen your awareness of how the information you’re encountering is affecting you, and how to stay true to your north star in a world of ever more persuasive AIs.
As the sculptor Constantin Brancusi once said: “Things are not difficult to make; what is difficult is putting ourselves in the state of mind to make them.” I’ll show you ways of reliably putting yourself into that state.
#3. Finding alpha
AI is transforming the information landscape, including which kinds of information are rare and valuable, and which kinds are abundant and cheap. For that reason, the choice of what content to consume and how is increasingly a key factor in whether your consumption habits distract you and lead you astray, or enrich you and amplify your knowledge.
I’ll introduce you to the concept of “alpha” – information that is novel, unique, or high signal – and how you can ensure you’re getting it. Together, we’ll map the landscape of your existing content consumption habits, identify a set of principles and filters you can use to increase the “alpha” in your information diet, and redefine the role of human notetaking in this era of abundant intelligence.
#4. Cloud storage for your context
Local file systems have increasingly become the de facto memory system for most LLMs this year, with the rise of “agentic harnesses” like Claude Cowork/Code and ChatGPT Codex/Work that can read and interact with local folders.
That approach has many benefits, but it doesn’t mean you can’t also gain the advantages of cloud storage, such as automatic version control, sharing and collaboration features, access from any device, and expandable storage space. The key is to sync the local folders on your computer with a cloud storage service, so that they mirror each other.
I’ll guide you through the minefield involved in doing so, including what to do if you have multiple accounts or run out of disk space, the best way to provide these folders as context to LLMs, and how to keep your cloud folders organized and easily accessible on desktop and mobile.
#5. Connectors – your AI’s nervous system
In the past, I was wary of connectors and recommended against them. But after spending a month going deep on every aspect of how they work, I’ve completely changed my mind and am now a big believer in them.
That is, as long as you approach them holistically, systematically, and with their gaps and shortfalls in mind. You have to know the pros and cons of connectors and MCPs versus using your local filesystem, how to handle team operations, and how they work differently depending on your device.
I’ll lead you through a process I’ve tested extensively to map your existing connectors, identify where your canonical data lives, vet its completeness and quality, and prioritize which ones to hook up first.
#6. Computer use
The tool known as “computer use” has come into its own in the last six months, led by ChatGPT. For the first time, AI has “hands” – that is, the ability to reach out and interact with the tools on your computer and in your browser, just as you would.
But as always, there is a lot more to using this capability effectively than any YouTube tutorial can convey. There are multiple ways of enabling LLMs to use your computer, each with their unique pros and cons and pitfalls.
I’ll help you understand the quickly evolving landscape of computer use across different LLMs, including which one to use depending on your use case, how to protect your security and privacy, and what you’ll need to set up upfront to allow AI to do valuable work without facing constant roadblocks.
#7. 12 Favorite Problems for AI
We’re bringing back and updating one of the core concepts of Building a Second Brain – your 12 Favorite Problems – which have suddenly become much more relevant in the AI era.
The right scope for AI-assisted exploration is not the project, the note, or the task – it’s an open question, encompassing a wide range of possible influences and answers that can be much more efficiently searched with LLMs. These are recurring questions that inspire your reading, learning, and growth over long time horizons, now with the help of AI.
I’ll take you through a process I’ve developed and refined to identify the most resonant and inspiring list of favorite problems, and set them up as a persistent system that your LLM can regularly revisit and research with minimal involvement required on your part.
#8. Routines (local and remote)
Another feature that has matured over the last 5 months is “routines” – regularly scheduled and recurring actions that LLMs can take on your behalf.
These routines can now function locally or remotely, resolving the main constraint on their effectiveness, which has been needing to keep your computer on and connected to the Internet.
I’ll explain the subtleties of how routines work, their best use cases, what other features need to be turned on, and how to set them up as part of a broader system for continuous learning and productivity.
#9. HTML artifacts
There’s been a sea change over recent months, from using markdown as the preferred format for all AI interactions, to HTML as an increasingly viable alternative for many use cases.
HTML is the default format of the Internet, and for that reason comes with a number of built-in advantages that are also useful for LLMs: higher information density, tables and graphs, visual clarity, headers and formatting, interactive elements, images and illustrations, tabs and links, mobile responsiveness, ease of sharing, and others.
I’ll teach you how to make use of all these strengths in your own workflow, even if you’re not a designer or web developer and have never written a line of HTML in your life. We’ll explore use cases that directly enable the projects and goals you’re pursuing now.
#10. Creating your own skills
In the last cohort, we provided a set of pre-built skills to help you build each component of your AI Second Brain. Skills have matured tremendously over the last few months, and there’s now a clear set of best practices for designing them yourself.
I’ll present a framework for thinking holistically and strategically about the human skills you may want to delegate to AI going forward, including having them interface with your favorite tools, setting up self-improving loops, and documenting everything you build in an accessible way so it doesn’t get forgotten.
You’ll notice that none of these features or tools are brand new. They’ve been around, in some cases, for quite a while. Our philosophy is that it pays to stand back a bit from the bleeding frontier, wait to see how things play out, and adopt new building blocks into your AI Second Brain only as they become mature and clearly useful.
Join a Free Live Session from The AI Second Brain
On Wednesday, August 26 at 12 pm ET, I’m teaching a free 90-minute session on the Master Prompt Method live.
This is the opening session of The AI Second Brain, and we’re making it available to everyone for free.
I will cover:
- How AI is already changing work, based on the latest research on jobs and skills
- The context window and its five layers, and why your leverage lives in just two of them
- Why “trained on” doesn’t mean “knows,” and how that reframes what people call hallucination
- A live walkthrough of the Master Prompt Builder, the practical tool we use to turn all of this into a working document
- Four ways to separate your personal and work context, so you can choose the setup that fits you
You’ll leave knowing exactly how to build your own Master Prompt, the master key to unlocking AI’s potential, and yours. We’ll send you the free builder and the full recording so you can put one together at your own pace.
We will also share the full details of Cohort 2 of The AI Second Brain and officially open the doors for enrollment.
Can’t make it live? Everyone who registers gets the recording.
Follow us for the latest updates and insights around productivity and Building a Second Brain on X, Facebook, Instagram, LinkedIn, and YouTube. And if you're ready to start building your Second Brain, get the book and learn the proven method to organize your digital life and unlock your creative potential.