MarsBit has published an edited write-up of a four-hour conversation between Sun Yuchen and Alan Shao, stripping out repeated points and casual chatter while keeping what the article describes as the most important and most counterintuitive takeaways.
The piece opens by sketching the two speakers. Sun is presented as a Peking University graduate who became a crypto entrepreneur, bought a lunch with Warren Buffett, went to space, and has remained at the center of wealth, controversy and technology waves. Alan Shao is introduced as an English-speaking content creator with tens of millions of followers and the person behind the “lean muscle theory.” The article says the 30-year-old sums up the last decade of his life in blunt terms: make money, invest, and reach financial freedom as quickly as possible. It adds that this stage goal has now largely been achieved.
Despite their very different backgrounds, the conversation starts from a question many young people would recognize: if you were 20 again, had no network, no resources, and only 1,000 yuan a month to work with, how would you begin?
According to the MarsBit write-up, Sun did not start by recommending a major or telling young people to bet on a specific track. He first laid out a personal “life operating system” built around three ideas: “Sun studies,” “lean muscle,” and “yellow-hair thinking.”
- “Sun studies” is about how a person understands wealth, freedom and opportunity.
- “Lean muscle” does not mean chasing oversized physique goals. It refers to keeping the body in a sustainable, low-burden, muscular state through strength training, diet and active health management.
- “Yellow-hair thinking” is about whether someone can keep executing when their choices do not match mainstream expectations.
Sun is quoted as saying, “Sun studies is the brain, basically an operating system upgrade. Lean muscle is a physical upgrade. Yellow hair is a mindset upgrade.” The article says that on top of those three layers, there is now a new helmsman: AI.
First, “AI-ify” yourself
The first major point in the article is that if someone were 20 again, the first move would not be making money. It would be turning themselves into an AI-ready system.
The write-up frames this through a practical question: if a young person has no computer, no savings, and only a phone, does AI still matter? Sun’s answer, as quoted in the article, is direct: “No computer? Everyone has a phone. Put it on the phone. It’s the same.” He also says, “No matter how much money a person has, anyone can do this for free now. First, AI-ify everything.”
In the article, “AI-ify” does not mean asking a chatbot a few questions or having it draft a piece of copy. It means bringing AI into the full workflow of life and work. Sun’s own method, the piece says, is to record diet, training, sleep and body data, save training videos, daily intake and exercise burn, and then have AI generate daily reports for ongoing tracking and adjustment.
He sums that up in one line: “Visualize it, datafy it, make it summarizable, then turn it all into Markdown documents.” MarsBit argues that the key point is not which model is used. It is the act of turning personal experience into data that can compound over time.
The article goes on to say that chat windows disappear, models change, and platforms may shut down. If the files stay local, though, they can be handed to Claude today and Codex or another model tomorrow. On that basis, Sun updates an earlier line of his — “Don’t chat on WeChat, chat with Doubao” — into a broader version: “Now upgrade again. Use APIs to talk to AI around the world. Keep all files local, then read them through APIs.”
MarsBit explains API in simple terms as an interface that lets different software systems exchange information. The point, it says, is not that someone must become a programmer first. The more important shift is to recognize that the data should stay in your hands, while the model is just a processor that can be swapped out. In that framing, the first asset worth building at 20 may not be savings, but a continuously updated personal archive that AI can call on.
Do not rush AI for answers before it knows you
The second theme is how AI should actually be used. The article says many people ask AI how to make money one day, how to study the next, and why they cannot stay consistent the day after. Every prompt starts from zero, so the output ends up sounding correct while remaining generic and not very useful.
Sun’s approach is slower and less intuitive. Record first. Do not rush to conclusions. He says, “Accumulate for 2 to 4 months first, then collect all the data and digitize all kinds of numbers.”
In the article’s telling, whether AI really understands a person depends on what that person feeds into it. Skills, income, spending, time use, health condition, projects completed, failed attempts, and what actually got done each day form the basis for any meaningful judgment. That is why, the piece argues, the real threshold for a personal AI assistant is not prompt engineering. It is sustained record-keeping.
The article also adds an important qualifier. Sun’s line about “100% trusting AI” should not be read as blindly doing whatever a model says. MarsBit says he asks multiple models at the same time, debates with them, and keeps tuning based on real-world results. A more dependable method, in the article’s summary, is to let AI widen the information set while reality handles final verification.
Young people may be deciding too early
On careers and life direction, the article says the bigger problem for young people may not be confusion. It may be locking in too early.
People are often told to find what they love quickly, pick a lane, and stay with it for years. Sun’s advice runs the other way. Do not decide first. Look at enough options first. He uses a housing example: “If you want to buy a home, look at 100 homes before making the decision. Don’t buy. Just look.”
The logic, as the article explains it, is that without enough samples, people do not even know what “good” looks like. After seeing enough homes, they can build a baseline for price, location, layout and negotiation. The same applies to career choices. If someone wants to choose a major, AI can be used to study how many majors may change over the next 10 years. If they want to enter an industry, they can compare barriers to entry, income, transferability and the likelihood of AI reshaping different roles. If they want to start a business, they can first study users, competitors and failed cases.
Sun uses a gaming metaphor for this process: map scouting. “Use it to scout the whole map first, mark all the resources that should be marked, then make the decision.” MarsBit says the worst player is the one who stays in the base producing units without scouting. That player does not know where the resources are or what the opponent is building, and only realizes the direction was wrong once the real fight begins.
The article says many people in real life “never scout the map.” They hear that a major pays well and go study it. They see someone else start a business and copy it. People around them take certificates, so they do the same. The problem is not a lack of effort. It is spending 10 years of effort to make up for 10 days of missing information at the start. One of AI’s most important uses, in this framing, is that it cuts the cost of scouting to a historic low.
Do not chase sectors. Look for “value breaks”
When Alan Shao asks what direction is most worth going all in on over the next few years, Sun mentions mathematics, astronomy, biology, pharmaceuticals, data services, AI and crypto.
But the article does not leave the point there. If all someone remembers is a list of sectors, it says, they will quickly fall into another cycle of trend chasing. Sun eventually brings business back to a simpler question: “At the most basic level, business is about whether you are creating value.”
MarsBit says the real opportunities often show up at several kinds of “break points”:
- Supply exists on one side and demand exists on the other, but the two are not connected.
- Large amounts of data have already been generated, but no one is processing them effectively.
- Expert knowledge exists, but it has not been translated into products ordinary people can use.
- Old processes are expensive and slow, while new technology can cut costs by 10x.
- A group of people keeps running into the same problem, but no reliable solution has appeared.
In that reading, entrepreneurship does not begin with a grand story. It begins with finding a broken link and connecting the two ends. If someone can help others get what they actually need at lower cost and higher speed, income follows as the result of value exchange.
Think globally, not only locally
The article also argues that what feels like extreme competition may simply reflect a market frame that is too small. Most people, it says, assume their customers are nearby, their market is their city, and their competitors are the same local peers.
Sun pushes back on that. “Business opportunities must be discovered globally,” he says. He adds, “Treat the whole world as your customer base. As long as you can show your value, you won’t have to worry about selling.”
MarsBit notes that China has strong supply chains, manufacturing capacity and talent density, yet many people still compete only inside a local demand pool. If 400 million young people are serving one another, competition will naturally feel intense. If the target is demand from 8 billion people worldwide, the scale of opportunity changes. The article says language, information and cross-border coordination used to block ordinary people from doing global business, but AI is weakening those barriers.
So the better question is not, “What jobs are left in my city?” It is, “What demand exists somewhere in the world that still is not being met cheaply, and which group of people can I create value for with the skills I already have?”
Freedom before wealth
On financial freedom, the article says Sun flips the usual order. People often assume money comes first and freedom follows. His line is the reverse: “Financial freedom is actually freedom first, wealth second. Wealth is a byproduct.”
In the article, freedom does not mean quitting work and doing nothing. It means keeping the ability to choose direction, location and lifestyle. Sun uses the phrase “live where the water and grass are,” saying, “Where there is water and where there is grass, that’s where we go live.” MarsBit interprets that as the ability to move toward places where value can be created, where new technologies appear, and where fresh demand and opportunity emerge.
That is why, the article says, Sun does not advise young people to lock themselves into one place too early through large mortgages, expensive consumption or rigid lifestyle commitments. The piece is careful to add that this is not a blanket rejection of homes, cars or marriage. The point is that every form of stability costs some degree of optionality. A major commitment may make it impossible to absorb a short-term drop in income. A fixed location may mean missing a new opportunity. A standard of living that must be maintained may make it harder to start over.
What matters more early on, in this framework, is building portable assets: work, skills, client trust, project experience, personal data and income sources that can move with you. Real freedom, the article says, is not doing nothing. It is not being forced to do the same thing forever.
As AI gets stronger, the body matters more
MarsBit calls this one of the most counterintuitive judgments in the conversation. Sun says, “In the AI era, everything will become very cheap. What is the only thing that stays valuable? Lifespan. So youth is wealth.”
The article explains the logic this way: if one person can direct dozens of AIs at the same time, the main bottleneck is no longer a shortage of assistants. It is whether that person’s body can stay online over the long run. That is where Alan Shao’s “lean muscle theory” comes in. On the surface it is about physique. Underneath, the article says, it is about active health management: not chasing exaggerated muscle size, but building a body that can support long-term work through strength training, sensible diet, sleep and body-fat management.
Sun uses two striking comparisons in the conversation: “You may be materially rich, but physically poor,” and “Muscle is the most important bank account in the body.” MarsBit notes that these lines do not need to be treated as strict medical conclusions. The underlying point is that the body is not something to manage only after success arrives. It is infrastructure for every long-term plan. Money, skills and models can all be updated. The body still has veto power.
The article extends that contrast by describing AI as a silicon-based team a person can mobilize, while the body is the carbon-based server that carries judgment and action. The stronger the silicon side becomes, the less room there is for the carbon server to fail early.
“Yellow-hair thinking” and Small Ego
Another section of the article focuses on two ideas that sound contradictory at first but are presented as complementary: yellow-hair thinking and Small Ego.
In the write-up, yellow-hair thinking sounds like ignoring advice and charging ahead. Sun defines it this way: “No matter what others say, if you believe you are right, execute fully.” The article says anyone choosing a nontraditional path will almost certainly face doubt. Many people do not miss opportunities because they fail to see them. They miss them because they cannot absorb the lack of understanding from people around them.
At the same time, Sun also stresses Small Ego. MarsBit says he uses Zhang Yiming as an example: “No matter how successful he is, he is always willing to hear what young people are saying and willing to learn new things. If something new appears, you have to learn it. If something new appears, you have to do it.”
The article’s conclusion is that yellow-hair thinking keeps a person from quitting because of ridicule, while Small Ego keeps them from freezing because of success. Together they form a more dependable execution style: stay firm when there is no new evidence, and change immediately when facts show you are wrong. The standard is not who speaks the loudest. It is what the evidence says.
Crypto as money rails for machines
On crypto, the article highlights one line from Sun that moves away from price action and speculation: “What we are doing is translating the money humans have used for thousands of years into a language AI and machines can understand. That is what blockchain and crypto are doing.”
MarsBit says traditional payment systems were built mainly for people and institutions. Crypto, by contrast, is trying to build a more open and programmable value network, where rules can be written into code, transactions can execute automatically, and funds can settle across a global network. That is also why crypto is often discussed alongside AI.
The article pushes the point one step further. If large numbers of AI agents eventually need to buy data, call APIs, rent compute or settle with one another, they will need more than a banking app designed for humans. They will need payment infrastructure that machines can call directly. MarsBit notes that readers may disagree with Sun’s view of crypto’s future, but says the question he raises is worth watching: if AI shifts from a tool that answers questions to an actor that participates in transactions and production, what kind of financial system will it need?
If machines cannot read it, it may as well not exist
The article also pulls out one especially forceful line: “If something does not become AI training material in the future, it is as if it never happened.” MarsBit says the wording is absolute, but the trend behind it is real. More and more knowledge will need to be digitized and structured before it can enter the next production system.
The piece gives several examples. In some industries, large amounts of know-how live only in the heads of veteran workers. In some teams, critical information is scattered across chat logs, spreadsheets and verbal handoffs. A person’s working method may disappear the moment they leave a job. Whoever can turn that experience into something machines can search, understand and call is doing foundational work for the AI era.
The article lists several entry points for ordinary people:
- build industry knowledge bases;
- turn repeated work into clear processes;
- structure cases, experience and exceptions;
- translate professional knowledge into products customers can use directly;
- keep publishing real practice so experience enters the network.
In that view, a future personal brand is not only about being known by other people. It is also about being legible to search engines and AI: what problems you have solved, what your method is, and what verifiable results you have left behind.
More experience, more creation
Near the end, the conversation moves from wealth, health and careers to a broader question: what is the meaning of life? The article says Sun does not offer a grand final answer. Instead, he describes himself as a player in “Earth Online,” someone who participates first, explores first, and does not rush to declare that he already understands all the rules at the start of the game.
He says, “I think experience is greater than conclusion,” and also, “The meaning of life itself may be the collection of your experiences.” MarsBit argues that this matters especially for young people. Many want guarantees before they begin: will this path work, can it succeed, will it become a lifelong career? But some information only appears after entering the field. You only know whether you like something after finishing a project. You only know what problems you are good at solving after serving clients. You only see what you missed before after entering a new environment.
If there has to be a broad direction for life, the article says Sun’s answer is simple: “Create more, consume less, and basically whatever you do will be right.” Creation can mean a service, a product, an article, a tool, or simply bringing order to a problem that has stayed messy for a long time. Consumption gives experience. Creation leaves a result behind.
MarsBit’s distilled takeaway
At the end, the MarsBit article compresses the whole conversation into a life operating system for young people: use “Sun studies” to rebuild how you understand wealth and opportunity, use “lean muscle” to maintain the body needed for long-term action, use “yellow-hair thinking” to resist outside noise, and use Small Ego to keep updating yourself.
Then connect AI to the whole system: record real data, open a larger map, look for value breaks on a global scale, and turn experience into assets that both machines and your future self can use.
The article closes on a direct note. Being 20 without connections or resources is not the frightening part. The frightening part is that the way the world works has already changed while a person is still using an old map to search for a new continent. Instead of rushing to ask which track can make fast money, the piece says, it may be better to become the kind of person who can keep finding opportunities, judging them, seizing them, and then has the body and mindset to carry the result through.