Original video in Chinese.
Key Takeaway
- I used Gemini 2.5 Pro to analyze and distill the core of my personal IP, with the goal of becoming a “super individual” and breaking free from dependence on the system.
- The two core levers for achieving a “super individual” are “code” and “narrative,” and they have enormous potential in the AI era.
- Through practice, I have worked out a way of learning and surviving in the AI era, emphasizing Learning by doing and Go fundamental.
- A personal AI system should have the ability to “navigate chaos,” not depend entirely on any single AI ecosystem, and keep creating products.
- This article shares a method for self-analysis using AI, understanding one’s own mental operating system through concept tables, relationship tables, and process tables.
Yesterday I had a very important conversation with AI: I asked the Gemini 2.5 Pro model to help me analyze and distill the core of my personal IP.
Over the past year and more, I’ve made more than 100 videos introducing all kinds of AI products, technologies, and methods. So, behind all these surface-level things, what exactly am I advocating for? Where does my driving force come from?
To be honest, I hadn’t thought out the answer to this question in advance, and I hadn’t summarized it over the past year and more either. So I came up with the idea of letting Gemini 2.5 Pro, with its massive context window, handle it for me.
In order to let Gemini understand my thinking, I exported all the scripts from my past 100-plus articles, over 75,000 words in total, fed everything in, and then carried out three rounds of conversation. Each round went deeper than the last. In the end, it produced a summary of three to four thousand Chinese characters.
To make that summary more readable, I then put it into the newly released Gemini CLI for visualization. And that produced this webpage.
I’ll walk everyone through it, and then you’ll know what I’m really sharing and what I’m really advocating for.
Start with the end in mind. The only goal I have in using AI is to become a “super individual.” Everything I share actually comes from my own needs, my own exploration, and my own practice.
Why do I have this idea?
Because after more than ten years in the workplace, I’ve never seen anyone end up well. I really want to get out of that game and break free from dependence on the system. So I want to become a “super individual” and hold the only means of production that is time in my own hands. In addition, I created the newtype community, screening rather than educating, to gather the minority like me.
So, how do you achieve that?
I chose two levers: code and narrative. If you want to change your fate against the odds, you have to use leverage. The classic levers are capital and labor. Unfortunately, neither of those suits me. As someone from a small town, the only things I can rely on are code and narrative. Especially in this AI era, I firmly believe that code and narrative are the levers that everyone—whether you’re a big name or a nobody—needs to grasp.
Along this direction, I’ve tried a lot of AI tools and methods. For example, building a personal AI knowledge base, methods for running local LLMs, and so on. I do all of this to achieve systemization and workflowization.
While practicing, I also figured out the way to learn in the AI era, or rather, the way to survive. For example, I’ve been emphasizing in the community that we should do Learning by doing, and not rely on so-called tutorials. We must go fundamental ourselves, and not blindly trust those self-media accounts, especially the domestic ones.
Taken together, all of this forms the shape of my operating system, including the kernel, drivers, applications, and firmware.
What I just said are all the Why—the reason I keep sharing AI content. If we dig one layer deeper, what else can we uncover?
I want to become an independent, self-realized “super individual.” But as an individual, I still can’t influence the AI ecosystem that I depend on so heavily. This is the contradiction Gemini uncovered.
So, when building my personal AI system, I will never depend completely on any single provider. For example, my knowledge base has gone through several iterations; it’s not just fully localized, but the model layer has also been separated out, so it can be swapped at any time.
Also, one major change I’ve made this year is that I’ve started creating products. Over the past year, I focused on internal optimization and shared my experience externally. This year, I’ve started making products myself. I use products to carry my methodology and values, and to output them externally.
So, digging further down from the Why, Gemini gave the final summary:
What I actually want to build is a personal operating system that can “navigate chaos”:
With the goal of achieving ultimate personal freedom, driven by code and narrative at its core, and with the strategy of mastering the high uncertainty of the external technology ecosystem, it is a living, continuously self-evolving personal operating system.
This is the core of my personal IP, and also the core of my personal cognition.
This time I only imported the video scripts I wrote. I actually also have notes, and I’m preparing to import them as well so Gemini can analyze them again.
If you usually accumulate a lot of personal material too, you can try this approach. The prompt is very simple:
If we compare a person’s thinking to an operating system, then this operating system contains three tables—concept tables, relationship tables, and process tables. So, based on the documents I provide, please analyze what is in my three tables.
This prompt is the first step: extracting the What from your materials.
Next, let Gemini further analyze and dig out the underlying philosophy, core principles, and ultimate goals that drive this system, which is the Why.
Finally, based on the earlier What and Why, go one layer deeper to identify the internal core contradictions, the path of evolution, and also predict the future challenges that may be faced.
If you do this whole big setup, I believe it will definitely inspire you a lot.
OK, that’s it for this episode. If you want to understand AI, want to become a super individual, and want to find like-minded people, come join our newtype community. See you next time!