Steer the Wheel or Stay Behind (click me)

This post is a synthesis of thoughts extracted from George Hotz’s blog. The 🐐 himself lately has been kinda hitting the spot, he seems to be talking directly to me as I have been stuck on overconsumption for I don’t even know how long now.

This topic, in all seriousness, is really close to my heart, and I have always felt conflicted about it. The thing is, even though me and George are very different people, I can assure you that we are both very similar in our insatiable curiosity, and I know that if you are a reader of this blog, you are too.
I have always considered curiosity a superpower, the true essence of existing. I don’t know if George still believes this (as a Christian now?), but I arrived at the conclusion that there is no real ‘meaning of life.’ There is no productivity or spending time right or wrong. The essence of existing is just spending time. Tigers chill below the sun, birds fly and rest—they are perfect in their ‘just existing.’ Humans created morality (which isn’t objective, btw) and, through their pretentiousness of creating and producing, ruined the world while at the same time improved it in incomprehensible ways.
This doesn’t mean that I am a nihilist and I run around killing people. I still judge my actions logically (don’t treat others how you don’t want to be treated) and I still highly weigh the fact that our time on this earth is limited—yes Bryan Johnson, I am talking to you too. For this reason, I value my time, my energy, and with that, what I consume.
Trying to consume from the right sources, trying to manipulate the algorithms to aid instead of hinder, to retain what I consume… I have been on this productivity shit (which is another self-loop of assurance and fake accomplishment), but I can’t lie that I fail most of the time. I gaslight myself that random consumption was a net positive in my life because it has led me into rabbit holes on the internet that have improved my conception of the world, making me who I am today. Yes, I believe that we are not only what we produce but also what we BELIEVE George. But on the other hand, I completely feel what George has been saying lately. Since I know that my natural tendency will always pull me towards consumption, I will have to try and live by what you will read below.

As always, I have been yapping. Let’s get the post started, which is partly AI slop—absolutely against what it will preach. You will love the contrast between what is written and the fact that it was partly AI-generated and a summary of another man’s thoughts… absolute comedy.
For this reason, below (as always) is a song that correlates with this post.

I. The Game You Were Handed vs. The Game You Should Play

Most talented people spend their entire lives playing a game they did not design, chasing rewards they did not choose, and wondering why they never feel like they are getting anywhere. The first decision an individual must make — before any other — is whether to steer the vehicle of their own life or to accept a seat as a passenger in someone else’s.

The person who chose the wrong game is easy to recognize in retrospect. They spent their years chasing total compensation, advancing in corporate hierarchies, accumulating titles, gathering prestige from institutions, surrounding themselves with the symbols of success that other people invented for them. The diagnosis is precise: they sought status instead of skill. Status is a positional good — it exists only relative to others, so every gain for you is a loss for someone else. Skill is not. Skill compounds on itself, transfers across domains, and produces real things in the world regardless of whether anyone ranks you for it. The person who spends a career building real technical ability will still have that ability when the company that employed them collapses. The person who accumulated titles will have a LinkedIn profile and a severance package.

The first actionable decision is therefore to conduct a ruthless audit: what exactly are you actually building in yourself? Is it the ability to do things the world needs done, or is it the appearance of being someone the current institutional hierarchy approves of? These are not the same thing, and the confusion between them is the source of most of the stagnation that talented people experience.

II. Consumption Is Not Participation

One of the clearest dividing lines in any serious examination of how individuals grow is the distinction between creation and consumption. The hostility to a pure consumer identity is not an aesthetic preference — it is a philosophical position grounded in a fundamental moral distinction.

Consumption versus creation is the distinction between childhood and adulthood. A child receives. An adult produces. This is not about age — it is about mode of engagement with the world. The tragedy is that an enormous number of adults have been trained, quite deliberately by the industries that profit from their passivity, to take pride in their identity as consumers. They define themselves by their preferences among existing products, by what brands they associate with, by what media they endorse and what entertainment they follow. This is not participation in the world. It is decoration of the self using other people’s work.

Consumption produces nothing. The money you spend consuming was earned somewhere upstream by someone who built something, and the act of consuming it does not regenerate what was spent — it depletes a resource and produces a culture that treats novelty of preference as a substitute for depth of understanding. As more people spend more of their time consuming, there is progressively less of value being produced and therefore less to consume. The system feeds on itself.

The practical reordering this demands is to treat your attention as a production input, not an entertainment budget. You spend it on things that compound — understanding that transfers, skills that generalize, problems that you are actually building toward solving — rather than on things that expire the moment the screen goes dark. Every hour spent consuming passively is an hour not spent building the internal substrate that makes you genuinely capable. If the balance of your life is consistently toward consumption, you are moving in the wrong direction regardless of how sophisticated or culturally validated your consumption happens to be.

III. Skill vs. Status: Why You Should Chase the Real Thing

There is a particular kind of understanding worth obsessing over: the kind you can verify, grounded in the ability to reconstruct what you claim to know. The distinction between knowing the vocabulary of a field and genuinely understanding the system it describes is radical, not semantic, and it has enormous consequences for how you should approach learning anything.

Consider two papers from different scientific traditions that make the same uncomfortable point. One asks whether a biologist could fix a radio. The other asks whether a neuroscientist could understand a microprocessor. Both arrive at the same answer: probably not, because the tools and frameworks of each field are tuned to describe systems, not to diagnose and repair them. To fix a radio you must understand electricity well enough to trace faults through an actual circuit under actual conditions. To fix a microprocessor you must understand logic gates well enough to follow actual signal propagation. Description and understanding are not the same thing. The rule generalizes without exception: you understand something when you can build it. You will know you understand life once you can build life. You will know you understand a brain once you can build a brain.

The surface markers of knowledge — the ability to repeat the right words, cite the right names, pass examinations — are exactly what institutional education optimizes for, and they are a completely unreliable proxy for real understanding. Real understanding is demonstrated by the ability to do something with the knowledge: to predict, to build, to repair, to generate something new. Anything less is performance.

The practical test is merciless: if your predictions are wrong, you do not understand. If you cannot build it, you do not understand. Your model of something must generate accurate predictions about how that thing behaves, or your model is wrong — regardless of how many books you have read about it. This is the only standard that resists the inflation of credentials.

For someone trying to retain and improve on what they learn, the consequence is this: the test of whether you understood something is whether you can produce something with it. Not whether you can recall it on demand, not whether you can discuss it fluently, but whether it is inside you as a functional capability rather than a stored label. Every time you study something, the question to ask is not “do I remember this?” but “what can I now do that I could not do before?” If the answer is nothing, you have stored noise, not knowledge.

The same mechanism plays out at the organizational level. People who play zero-sum games inside institutions — tracking their seniority, managing their internal brand, optimizing for appearing valuable rather than being valuable — actively damage the teams they join. They consume organizational trust and produce political complexity instead of solutions. When your motivation for improving is comparative, when you are learning in order to rank higher rather than to solve real things, you produce the same outcome inside your own mind. You are engaging in a zero-sum performance. The person who learns in order to solve real problems retains things differently, uses them differently, and compounds differently.

IV. The Trap of the Zero-Sum Mindset and Its Internal Equivalent

A large and growing portion of professional life is spent on zero-sum redistribution rather than genuine value creation. These are people who produce no real output but who successfully redirect flows of value toward themselves through political skill, credential accumulation, and status games. Estimates of how many people fit this description range between a third and three quarters of the working population, depending on how honestly the question is asked. The precise number matters less than the recognition of the pattern: an enormous amount of human effort, including talented human effort, is being spent on fighting over the existing pie rather than making it larger.

What is less obvious is that this is not merely an economic or societal phenomenon. It is a psychological one you can succumb to entirely inside your own head. When your motivation for improving yourself is comparative — when you are learning because you want to be better than someone else, when you are building skills in order to rank higher in some hierarchy, when you are working harder in order to consume more status symbols — you have internalized the zero-sum mindset. You are, in your own inner life, the parasitic class: redistributing effort toward appearances and away from genuine capability.

The antidote is stated simply: do not play zero-sum games. Create value for others and do not obsess over the returns. If you create more value than you consume, you are welcome in any well-operating community. The freedom this creates is significant. If you are not competing for position, you cannot fall behind. You are growing a capability that belongs to you absolutely, not a rank that exists only relative to others and evaporates when the ranking system changes.

There is a famous experiment with monkeys and food. Give each monkey a slice of cucumber and they are all perfectly happy. Then give one monkey a grape — a clearly superior reward — while the others continue to receive cucumbers. The cucumbers have not changed. Nothing about the monkeys’ absolute situation has changed. But they now throw the cucumbers back at the experimenter in rage. The grape destroyed the value of everything they had.

This mechanism runs constantly in human motivation. People abandon what they have and what they are building the moment they glimpse that someone else has something better. The solution is not stoicism about the grape — it is the deeper recognition that the cucumber is real and the ranking system that makes you feel bad about it is constructed and arbitrary. The grape exists in the comparison. Remove the comparison and you still have the cucumber, which is food, which is good, which is what you came for.

V. You Have to Actually Build Things to Understand Them

The distinction between understanding something by building it versus understanding something by reading about it or thinking about it abstractly is the most important distinction in all of learning, and it is the one most consistently bypassed by people who are nominally serious about improving.

Consider what it means to use an AI model to write code. The model takes a description in English and outputs something that may work. If you do not understand the underlying system well enough to evaluate the output, to debug it when it fails, to extend it when requirements change, you are not a programmer. You are someone who has outsourced understanding to a tool and is now permanently dependent on that tool. The moment the tool fails or is unavailable or produces something subtly wrong in a way you cannot detect, you are helpless. You never had the understanding. You had the output.

This is the correct frame for any form of learning. The lecture, the textbook, the blog post, the conversation, the summarized note — these are all forms of receiving a compiled output. Someone else has done the compression, the organization, the selection. You are reading their map. Reading a map is useful, but you do not know the territory until you have walked it. The map and the territory are not the same thing, and the difference becomes brutally apparent the first time you need to navigate somewhere the map did not cover, or the first time the map was wrong.

The actionable practice is to always close the loop between learning and building, even at small scale. If you read about a concept, build something with it, even something trivial and throwaway. If you watch a tutorial, immediately construct a toy version of what you just watched. The act of construction reveals everything you thought you understood but did not. This is not about productivity. It is about epistemology — about how you actually know whether something is inside you or merely near you. The difference between inside and near is the difference between understanding and familiarity, and familiarity without understanding is an expensive illusion.

VI. The Architecture of a Mind That Actually Grows

There is a useful framework from machine learning that, when translated into terms of human cognition, describes the architecture of a mind that genuinely compounds versus one that merely accumulates.

Any system that learns from experience has three essential functions. The first is representation: taking in raw sensory data and compressing it into a form that is actually learnable, discarding irrelevant noise and preserving what matters. The second is dynamics: maintaining a model of how things change over time, integrating the past and extrapolating toward the future — the equivalent of a mental model of how the world moves, not just how it currently is. The third is prediction: using the current state of knowledge to decide how to act, not merely to describe what has been observed.

Most people’s learning lives are dominated by the first function and weak in the other two. They accumulate information — take in facts, concepts, data points, ideas — but they do not build a robust dynamics model that allows them to predict how their domain is evolving, and they do not develop a prediction function that lets them use their model to act differently than they would have acted without it. The person who reads widely but cannot predict anything, cannot make better decisions, cannot produce better work, has representation without dynamics and no prediction function at all. They are a library without a search engine — full of information that cannot be retrieved when needed and cannot be acted on when the moment demands it.

The practice of building a dynamics model means asking, constantly: given what I know, what should happen next? And then checking whether it does. The habit of predicting — about your field, about the projects you are working on, about the behavior of systems and people you interact with — and then tracking whether your predictions were accurate is the most powerful retention and growth mechanism available, precisely because it is painful when you fail. Cognitive dissonance, the friction between what you predicted and what actually happened, is the signal that the model needs updating. Every failed prediction is a data point you cannot get from passive reading, and it is worth far more than a successful one because it tells you exactly where your model is wrong.

VII. The Entropy Problem and Why You Must Fight It Actively

Physical systems decay unless energy is continuously applied to maintain them. A house left alone does not stay a house — it becomes rubble on a long enough timeline. A pump left unserviced fails. A road left unrepaired returns to dirt. Entropy is the default state of everything physical, and fighting it requires constant, deliberate effort applied without cease.

The same is true of knowledge and capability. Information that is not used decays in its accessibility. Mental models that are not tested drift away from reality. Skills that are not practiced rust into vague memory. The person who stops building, stops predicting, stops constructing new things watches their genuine capabilities erode even as their credential count stays constant. A degree does not expire on paper, but the understanding it certifies absolutely expires if it is never renewed through practice. The surgeon who has not operated in five years is not a surgeon in any meaningful sense. The programmer who has not written code in three years has lost more than they think. The knowledge is not gone — but the capacity to deploy it fluidly, accurately, under pressure, has degraded below the threshold of usefulness.

The active response to entropy is maintenance through use. Not review for its own sake — not passive re-reading, not flashcard drilling as an end in itself — but genuine deployment of what you know against real problems that punish you for getting them wrong. The engineer who writes code every day, the mathematician who works through new problems every week, the strategist who makes explicit forecasts and tracks them, is fighting entropy by putting knowledge under load. The person who attends conferences and reads about things and discusses ideas is performing the activity of someone who knows things without actually keeping the knowledge alive.

The implication for genuine independence is direct. The person who is genuinely capable — who can represent, model, and act accurately in their domain — is much harder to displace than the person whose value is purely credentialed. Credentials can be devalued by institutional changes, by shifts in what institutions consider relevant, by the simple passage of time. A deep, embodied technical understanding cannot be confiscated, cannot be inflated away, and does not depend on any external authority to remain valid. It is yours the way physical strength is yours — maintained by use, lost through neglect, but not contingent on anyone else’s permission.

VIII. Do Not Accumulate the Map, Accumulate the Territory

Money is a map. This is not a metaphor — it is a precise statement about the nature of currency in any economic system. The map is supposed to correspond to the territory: to represent real economic value, real things being built, real problems being solved, real people being genuinely served. When the map corresponds accurately to the territory, it is useful. You can navigate with it. But when the territory changes and the map does not, or when people begin optimizing for the map itself rather than for what the map is supposed to represent, the map becomes actively misleading.

When one entity accumulates enough of the map — when money concentrates sufficiently — the money becomes worthless, because there is nothing left of the territory on the other side of the transaction. You cannot buy things from people who have nothing. The market requires participants on both sides. This is not a moral argument against wealth — it is a mechanical description of how the system fails when people confuse the representation for the thing being represented.

Translating this to the individual level: accumulating money as a hedge against an uncertain future is the wrong strategy precisely because the institutions that denominate that money have their own interests and will find ways to protect those interests at the expense of people who merely hold the currency. In a world where the relationship between capital and labor erodes — where machines increasingly do what workers used to do — holding the currency of the old system buys you diminishing amounts of what you actually need. A pile of money will buy you nothing in a world where the people who control what money can purchase have decided not to sell to you.

What does buy you something is the kind of understanding that lets you produce value directly, without needing an institution to validate your output and pay you for it. The person who can write code that runs, design systems that work, build tools others actually use, does not need the permission of a credential-granting body to demonstrate their value. Their value is demonstrated by the thing itself. The thing runs or it does not. The system works or it does not. There is no committee required.

Every hour you spend learning something that only has value inside a specific institutional context — a particular company’s internal processes, a certification that only matters within a specific industry’s hiring culture, a credential that signals things to HR departments but does not make you more capable of doing anything — is an hour invested in the map. Every hour spent learning something that makes you more capable of producing things directly, of building and solving with less dependence on external validation, is an hour invested in the territory. The territory outlasts every map that has ever been drawn of it.

IX. The Rhetoric of Fear Is Not a Strategy

Social media has developed an extraordinarily efficient mechanism for generating anxiety and routing it toward engagement. The mechanism is simple: there is always someone who appears to be doing more, building faster, adopting tools sooner, achieving more by any visible metric. The algorithm surfaces these people specifically because the gap between what you are doing and what they appear to be doing generates anxiety, and anxiety generates engagement, and engagement generates revenue. You are the product. The fear is the product.

This is why every new technology arrives accompanied by a chorus of voices insisting that if you do not adopt it immediately and completely, you will fall permanently behind and become economically worthless. This rhetoric reliably gets enormous traction because it is targeting a real anxiety about the future. But the claim itself is almost always false. Real improvement is slow, compound, and unglamorous. It does not require you to adopt every new tool the moment it appears. It requires you to build genuine understanding of a small number of things deeply enough that you can actually use them. The person who genuinely understands one system and can build things with it is far more resilient than the person who has superficially adopted twelve tools because an influencer said each was mandatory this month.

Artificial intelligence is a useful illustration. It is not a magical game-changer. It is the continuation of a long exponential of progress that has been underway for decades. It is a win in some areas, a loss in others, and overall a useful tool — like a compiler that takes imprecise input and produces code, with all the limitations that implies. A compiler replaced certain kinds of programming jobs. Spreadsheets replaced certain kinds of accounting jobs. AI will replace certain kinds of knowledge work, in the same incremental and domain-specific way, not in the sudden recursive explosion the most excitable voices promise. The study showing that AI assistance makes developers feel twenty percent more productive while actually making them nineteen percent slower is precisely the kind of result that does not get amplified, because it does not generate the anxiety that drives engagement.

The individual response is to develop a strong prior that any claim of the form “if you don’t do this immediately you will fall behind” is marketing, regardless of who is making it. This does not mean ignoring new tools. It means integrating them as tools rather than as identities, after understanding them deeply enough to know what they actually are.

X. What It Actually Means to Produce More Than You Consume

The simplest ethical framework that holds up under scrutiny is this: at the end of your life, total it all up. You should have produced more than you consumed. The fry cook who feeds people every day is a good person by this metric. The billionaire who builds companies, employs people, and solves problems is a good person by this metric. The private equity manager who extracts value from functioning companies without creating anything is not — regardless of how much money they accumulated in the process.

The standard is not output size. It is net direction. Are you, on balance, adding to the world’s stock of things or drawing from it? You should not feel bad for producing less than you could in a given period, for living off savings for a stretch, for being in an investment phase that will pay off later. The standard is lifetime net, not quarterly. The problem is when consumption becomes the identity and production becomes the exception, when the investment phase never ends and the deployment never begins.

There is also an important clarification about who the actual problem is in any society. It is not the immigrants or the billionaires, the targets that political rhetoric reliably identifies because they generate the strongest emotional response. The actual problem is the unproductive class — both rich and poor — the finance middlemen and the rent seekers, the people who purposefully create complexity for others, the lobbyists, the obstructionists, anyone who makes their living by taking from the productive rather than by producing alongside them. The solution is not to tear down the productive — it is to stop tolerating and elevating the extractive as if their activity were a legitimate contribution to the whole.

For someone trying to improve and retain what they learn, this framework has a direct application. Every period of input — reading, studying, watching, absorbing — should be in service of a period of output. The learning that does not eventually produce something is consuming the resources of the learning that does. Invest in yourself for a period, absolutely. But if you are permanently investing and never deploying, the investment has not landed. It is sitting in your head as latent potential that you are never converting into anything real, and latent potential subject to entropy decays into nothing.

The cycle is: take in something real, build something real with it, note where your understanding failed, take in what you need to fix the gap, build again. This is not a productivity framework. It is a description of genuine growth — the kind that compounds, that resists entropy, that does not require anyone else to validate it because the things you build validate it for you.

XI. The Steering Wheel

There is a question that cuts through almost every problem of individual growth and self-direction: why try to beat them when a million others tried? What would have to be different for any of this to actually work? What do you have available to you that everyone who attempted this before did not?

The answer is always the same: technology changes what is possible. The tools available now are not the tools available a generation ago. The paths that were blocked before are not all blocked now. The question is not whether old approaches failed — they did, reliably, for most people who tried them — but whether you are clear-eyed and creative enough to find the paths that are genuinely open rather than the ones that look open because everyone is crowded onto them.

The steering wheel is not a metaphor for working harder. It is a metaphor for taking actual control of the direction of your life — examining what game you are playing, who designed that game, whether winning it gets you anywhere you actually want to go, and whether there is a different game you could be playing instead. The passenger in someone else’s vehicle can be very busy. They can be productive by every internal metric the vehicle provides. They can accumulate a great deal along the route. But they do not get to choose where the vehicle goes, and when the driver decides to go somewhere else, the passenger goes there too.

The individual who takes the steering wheel — who builds real things, who values skill over status, who creates more than they consume, who fights entropy through genuine practice rather than credential accumulation, who resists the zero-sum mindset and builds instead toward genuine value — is the individual who ends up somewhere. Not necessarily somewhere comfortable. Not necessarily somewhere prestigious. But somewhere real, built by their own hands, understandable in terms that have nothing to do with what other people decided to value this quarter.

Build the world you want to see tomorrow. Not because it matters in some cosmic sense. Because you can. And what the hell else are you doing with the hours?

Thanks @geohot, for everything.