Feedback wanted. Email me!

What's needed? What's missing? What else?

The (incomplete) Linking Your Thinking Manifesto on AI

AI is wonderful and terrible. It can do more and more for us. But there are some things that it cannot, and should not, be replaced.

This manifesto is not about what AI might be able to achieve in the hard sciences—the physics, maths, and other areas where artificial computational power is beneficial. It’s also not about the many uses of AI for productivity gains, such as transcribing and summarizing meetings, building technical solutions for non-technical people, doing preliminary research, and other uses of a similar nature.

This manifesto is about our personal relationship to AI—and how that relationship has the best chance of remaining healthy.

Stop optimizing for lowercase “u” understanding

“I just threw 50 sources into NotebookLM and generated an AI podcast of the material.” Cool. But did you actually listen to the entire 20-minute podcast?

For the two people left standing in the room, do you actually think you know the material now? Are you talking intelligently about it the next week? Can you tangibly apply what you “learned” in your projects and pursuits? The answer is “no,” but AI makes us feel like it’s “yes,” and that’s a dangerous conceit. It inflates a false sense of “I know this now,” that can stunt personal growth and true, capital “U” Understanding.

AI is great at helping us acquire information faster. It’s fantastic to have an AI assistant handy to readily answer a few of those pesky 1,000 ideas some of us seem to have throughout the day. AI is great at helping us rapidly understand the contours of a query or topic. But that type of understanding is a lowercase “u.” Let’s call this what it is. Superficial learning.

Superficial learning is not all bad and can be accelerated with use of AI. For many situations, quickly “getting a handle” on a topic is quite valuable. AI is wonderful for that. However, superficial learning is superficial: existing or occurring at or on the surface.

Until AI advances to the point where it can not only rewire our neurons, but also add emotion, memories, and even affect the nervous system—it will never be able to do the work for us. And even then, we are missing the point. Much of the time, the work of understanding is the point of understanding—and understanding can’t be fully delegated. The journey is the destination. Simply arriving at every destination, seems great, until reality sets in that human striving is tethered to human happiness.

That’s why “Linking Your Thinking” isn’t only the name of our business. It’s what were optimizing for. “Linking Your Thinking” is a call for you to be the one making relationships between ideas. When we make the connection ourselves, that’s what most powerfully and lastingly makes the change in our brains and bodies.

AI can offer understanding in the form of suggestion, but we forge it ourselves through connection.

Voluntary Laws of AI Interaction

Don’t break the covenant. Don’t pass off AI writing as your own: You contribute to inauthenticity in the world. You prioritize influence and attention over authenticity. You break the covenant.

As Karen Hume writes, “There is an unspoken covenant between a reader and a writer. When that trust is broken, when I’m not sure that what I’m reading is coming directly and unfiltered from you, I’m not going to read you” to which I’ll add, “and I’ll think less of you.”

Passing off AI writing as your own is nothing less than a lie. An insidious one too, as we’ll argue in the next voluntary law.

Don’t let AI write for you. Hume continue to illuminate the other hidden cost to AI writing for you:

MORE TO COME IN THIS SECTION

THEN, WILL BE AN INVITATION TO USE "IDI" AS A MODEL FOR ENSURING HEALTHY ENGAGEMENT WITH AI.

Further incomplete notes on our relationship with AI

Hone your discernment so that you can trust your intuition. There are countless ways we might interact with AI. Each of us is different, with different needs and life experiences. No laws or rules could possibly account for all the edge cases of interactions that we humans will have now and into the future. Very often, you will come up to the threshold. Will you break the covenant for a quick buck?

If you don’t have skin in the game, it’s slop.

Breaking the covenant damages your virtue. Do the ends justify the means? Well, the means shape who you become. It’s circular but true, we become what we do. Even more, the means shape how people see you. Your followers will grow, and with them comes a form of power, but your peers, whose opinions may matter more, will think less of you—not because they want to, but because you gave them little choice.

MORE TO COME

Having produced multiple YouTube videos, an entire course called "AI + PKM," and an upcoming high-level bootcamp on creating your own semi-autonomous agent, Linking Your Thinking is not avoiding AI, nor are we blindly enthusiastic about it. We believe there is a middle path that keeps the human in the loop. While the middle path, outlined in this manifesto is voluntary, we believe it's necessary in leading a proud, virtuous life where making meaning is prioritized.

Are you going to the PKM Summit?

The 3rd annual PKM Summit is happening next month in Utrecht, Netherlands. I've gone the past two years and both times delivered on all levels, and this year, in collaboration with the organizers, I'm happy to announce a 10% discount to the newsletter! Register at 10% off for the PKM Summit here.​

Now, when I get off the train at Utrecht, I immediately feel at home. Utrecht has many warm memories; I really love this event. You meet so many great people. The energy is amazing. I hope you make it this year.

Here are some pictures from last year.

I have a lot of folders. But I don't need them, do I?

How many folders in Obsidian do you have. I have 360. And people think I don't like folders! But I don't really need them. Give me half a day and I bet you I could get down to 3. I might even find that I've improved my ideaverse. Why? Because then I would be maximizing the LYT mantra to organize for context.

How do you organize for context?

(And how many folders do you have?)

AI + PKM Bootcamp

Create a semi-autonomous AI employee and supercharge it with a shared Obsidian ideaverse.

In this one-time only, one-day bootcamp, you will leave with:

• Your own AI employee. A semi-autonomous agent that can work while you sleep.
• A shared Obsidian workspace ready for any AI.
• The exact documentation, guidance, and language you need for any AI to excel working for you.

This bootcamp is specifically for the Producer-driven archetype. Inner Guides and Creatives won't get much out of this. Synthesizers might.

We are targeting either:

• Thursday, April 2nd, 2026
• Tuesday, May 12th, 2026
• One day, seven hours

A few examples (of many) of what you can do when you combine an AI Employee with a shared Obsidian ideaverse.

• Hey AI, make a map of content with every note where I mention “Architects and Gardener.”
  • Get a new note with links to every mention.
• Hey AI, transcribe this two-hour long video with different types of takeaways at the top, and the full transcript at the bottom.
  • Get a new note with sections on: Key Themes, Live Exercises, Notable Quotes, Community Highlights (if there was a chat window), and the Full Transcript.
• Hey AI, help me create a new website.
  • Get an always-available collaborator guide you with setting up a domain name, nameservers, a hosting platform, a github repo (if needed)—along with the nuanced visual design of your website.

If you are interested in attending this one-time only event, which date is better...

In one day, get a semi-autonomous AI employee, supercharged with a shared Obsidian ideaverse, and all the setup, considerations, and documentation you need.

​If you are interested, make sure you fill out our brand-new, updated survey.

It asks about Claude Cowork, OpenClaw, and other pertinent questions.

(And yes, everyone on Knowledge Accelerator, gets this included.)

Idea Updates

Soft Paper is becoming an official Obsidian theme

I spent more time than I'm happy about taking the AnuPpuccin theme, which was beautiful but bloated and caused slowdown for a lot of people, myself included—and I cut out all the extraneous settings, until all that was left was Soft Paper.

The result? A theme that is 86% smaller and no longer slows down. It's not available yet, but here's a sneak peek.

You might notice how awesomely compact the bases views are in the sidebar. That's an option you can toggle. Personally, I LOVE it.

LYT Workshop - Week 3

We went over Idea Emergence, how to actually switch between architect and gardener, and we spent extra time with techniques to make the mapping method come alive. I also brought back the concept of STIR, the four elements of memory and retrieval, reimagined the slides and the timing of it, and it really landed.

As opposed the the LATCH model, STIR accounts for links. (But that's too much theory for today.)

Idea Exchange

​The Lascaux Cave paintings and how sharing is a human tradition are a cornerstone in the final unit of the LYT Workshop curriculum. It's awe-inspiring how so many added to the story some 17-22,000 years ago. While it doesn't always have the gravitas and longevity of those beautiful cave drawings, we can each add to the story with our shares. However you do it, keep sharing.

​The em dash. 99% Invisible has a great podcast episode on the em dash, which despite it being a false positive for using AI-assisted writing, is a staple in my personal style.

Stay connected,

Nick

P.S... Really want your Obsidian fix? Here are some quick-hitting shorts we put out this week.

Short 1: New to Obsidian? Make this note first.​

Short 2 How to use folders in Obsidian the right way.

To respond to this newsletter, just hit reply. I love getting replies, and read all of them, but I have sadly come to the conclusion I can't realistically reply to most. Trust me, I hate this. But, well, Life puts limits on all of us. Thank you for your understanding. (And if you received this email from a friend, and would like ​to subscribe, please go here​.)