The Great Inversion, in One Chart

There's a chart I keep redrawing in my head, and I want to finally put it on paper. Two lines. One is the value of doing the work — writing the email, building the deck, running the numbers — and for most of history it sat high because doing the work was hard and scarce. That line is falling, and the line underneath it, the value of judgment, is rising to meet it. The crossing is the part worth talking about.

The two lines on the chart

Draw it with me. Horizontal axis is time. The first line is the value of doing the work itself. Producing the thing. For thousands of years this line lived near the top, because production was the bottleneck. If you could draft the contract, set the type, balance the ledger, build the cabinet, you got paid for the doing. Skill at the task was the asset.

That line is dropping. A machine now does most of the doing in seconds, and it gets cheaper roughly every month. This is not a forecast. You have already felt it in the work that used to eat an afternoon and now takes a prompt and a once-over. A first draft. A summary of forty pages. A function. A passable image. The marginal cost of competent output is sliding toward zero.

The second line is the value of judgment. What to build. What to refuse. What is actually worth your Monday morning. That line is rising. Where the two cross is the moment I keep coming back to. The thing that used to be scarce — doing — became abundant. The thing everyone assumed — deciding — became the scarce one. That crossing is the Great Inversion, and on most desks it is happening right now, quietly, while the headlines argue about something louder.

Why abundance on one side creates scarcity on the other

The mechanism is simple, almost boring, which is why it holds. When the doing gets cheap, the deciding gets expensive. Those two moves are linked, not coincidental.

Think about what cheap output actually does. It removes the natural filter that effort used to provide. When a report took two days, the cost of producing it forced a decision before you started: is this report worth two days? The labor was the gatekeeper. Most bad ideas died at the gate because nobody wanted to pay for them in hours. Now the gate is gone. You can generate ten reports, forty subject lines, six strategies before lunch. The constraint moved. It is no longer can I make this. It is which of these was worth making, and which one do I send.

Naval Ravikant said it plainly years ago: in an age of near-infinite leverage, judgment becomes the most valuable skill. The chart is just that sentence drawn out. When everyone can produce, the rare and paid thing becomes knowing what should be produced. Volume is free. Direction is not.

There is a second-order effect worth naming. Cheap output also means cheap wrong output, at scale and at speed. A bad decision used to cost you one slow project. Now it can cost you a hundred fast ones, all polished, all confidently produced, all pointed at the wrong target. The cost of poor judgment did not stay flat. It multiplied with the leverage.

Why "learn the tool" is the wrong panic

The common reaction to all this is to learn the tool. Take the course. Master the prompts. Become the person on your team who is fastest at making the machine produce. That instinct feels safe because it is concrete, and it is not useless. But it is defending the falling line.

Productivity at the doing is exactly the thing depreciating. If your edge is that you can generate output faster, you are competing in the one arena where the price drops every quarter, and where your advantage over a colleague who learns the tool next month is measured in weeks. You are sprinting down the line that is heading for the floor.

The other failure is subtler and more common among capable people: drowning in your own production. When making things is free, the volume of things explodes. More drafts, more options, more dashboards, more meetings to review the drafts and options and dashboards. People mistake this motion for progress. They feel busier than ever and move the actual business less than before, because they trained the cheap muscle and starved the expensive one. The honest question stopped being can I do all of this. By hand, the answer is now no, and you no longer need to. The harder question is which of this deserved to exist at all — and that question does not get answered by being faster.

How to train the muscle that's rising

Judgment is not a personality trait you either have or lack. It is a muscle, and it trains under load. Here is how I work on mine, concretely.

Start each week by writing down the one decision that, if you got it right, would make the other forty decisions smaller or unnecessary. Not your task list. The upstream call. Most people never name it, so they spend the week executing around a question they never answered.

When you use the machine, change your role on purpose. Let it produce three versions, then spend your real time on the part it cannot do: choosing, killing, and stating why. Write the why down in a sentence. The reasoning is the asset, not the output. Over a few months those sentences become a record of how you actually decide, and you start catching your own bad patterns.

Practice subtraction. Once a week, look at what you produced and ask what should not have been made. Cutting trains the same muscle as choosing, and it is faster feedback. And protect thinking time like it is a meeting with your most important client, because it is now the highest-paid thing you do. At MARSA this is the whole posture we build toward — hand the doing to the machines, and put the freed hours into the deciding. The leverage is only worth having if a clear human is holding the other end.

The objection: doesn't taste collapse too?

Here is the fair pushback. If the machine keeps getting better, won't it eventually out-decide us too? Won't judgment itself get automated, and this whole chart just delay the same fate by one step?

Maybe, at the narrow edges. A model can already weigh options inside a defined frame better and faster than most people. But notice what it cannot do, and why. It does not know what you are for. It has no stake in the outcome, no taste built from your particular failures, no sense of which trade-off you can live with at 2 a.m. and which one will quietly rot you. Judgment is not only computation over options. It is computation in service of values someone actually holds. The machine can rank the menu. It cannot be hungry for you.

There is also the accountability piece, which people skip because it is unglamorous. Decisions carry consequences that land on a person. Someone has to own the call when it goes wrong, learn from it, and carry that learning into the next one. That loop — choose, live with it, update — is how judgment compounds, and it only runs in someone who bears the cost. So the honest version is not that humans stay supreme forever. It is that the scarce, paid skill keeps being the one closest to wanting something and answering for it. For now, that is still us.

The quieter life on the other side of the line

Step back from the chart and the shift is gentler than the panic suggests. For most of history we were valued for our hands and our hours. The work used us. The Great Inversion does something strange and a little hopeful: it makes the most human part of us — what we choose to care about — the part that pays.

That is not automatic comfort. A lot of people built their whole identity on being the one who does the work, who stays late, who grinds it out. Watching that line fall feels like watching your worth fall, and I do not want to wave that away. The grief is real. But the thing rising in its place is more yours, not less. Nobody can out-decide you on the question of what your life and your work are actually for, because nobody else is living it.

The move, then, is not to cling to the doing or to fear the machine. It is to let the doing go where it is going — cheap, fast, handled — and to deliberately become the person at the other end of all that leverage. Clear about what matters. Slow to add, quick to cut. Awake to the decision underneath the tasks. The chart is just a picture of where to point your attention now. One line down, one line up, crossing about here. Stand on the line that is rising.