What Is the Human For? The Thesis Behind Everything

The machines learned to do the work almost overnight, and most people felt the floor tilt. I didn't read it as a threat. I read it as a question that had been waiting a long time: now that the doing is cheap, what is the human actually for? This is the answer I keep arriving at, written down in one place.

It was never the circumstances

Work with enough people and the same thing keeps surfacing. Two founders with nearly identical companies, same revenue, same market. One is steady. The other can't sleep. Two people on the same diet, same training plan, same starting weight. One holds it for years. The other quits by March. The variable isn't the plan. It isn't the bank balance or the calendar or the genetics, though those matter at the edges.

What differs is the thing running underneath. How they pay attention. How their body handles a hard week. The story they tell about what a setback means. The habits that fire before they've decided anything.

We're trained to look at outcomes and assume the outcome was the problem. So we change the circumstance. New job, new city, new program, new partner. And the relief is real, for a while. Then the old pattern reassembles in the new setting, because the pattern was never about the setting. It traveled with the person. I've watched this happen too many times to call it coincidence. The circumstances are downstream. The system is the thing.

The system underneath a life

When I say system I mean four things that are doing most of the steering, mostly without your permission.

Attention decides what's even real to you. Your brain isn't a camera taking in the world; it's a prediction machine that spends most of its energy guessing what's coming and filtering for evidence. What you habitually attend to becomes what you expect, and what you expect shapes what you notice next. Aim it at threat and the world fills with threat.

Stress is the second. Short bursts are useful and clarifying. The problem is load that never resets — the nervous system stuck in a low hum of alert. Hold that long enough and it costs you sleep, digestion, mood, and judgment. The body keeps the tab open.

Story is third. The same event — a lost client, a hard conversation — gets a meaning before you've consciously decided one. Psychologists call this appraisal. The appraisal, more than the event, drives what you feel and do next.

Habits are fourth. Most of a day runs on routines the deliberate part of you never signs off on: a cue fires, the behavior runs, a small reward locks it in.

Change these four and the life changes, in a way new circumstances never quite deliver.

Why the usual fix doesn't hold

The standard playbook is willpower plus more output. Try harder. Add discipline. Do more, earlier, faster. It works for about three weeks, which is roughly how long motivation lasts before it returns to baseline.

Here's the mechanical reason it fails. Willpower draws on the same deliberate, effortful system that's already exhausted by a stressful life. You're asking the most expensive part of the brain to override the cheapest, most automatic part — and the automatic part runs on default. Under fatigue or pressure, the deliberate system goes offline first. That's not weakness. That's how the wiring is built.

The output trap is subtler. We measure ourselves by what we produce, so when something's wrong we produce harder. But output was never the scarce thing, and it's now genuinely cheap. A person grinding out more work to fix a dysregulated system is bailing water without finding the hole.

The fix that holds doesn't fight the automatic system. It edits it. You change the cues, lower the baseline stress, redirect attention, and rewrite the story — so the easy, default behavior becomes the one you actually want. Less force, more design.

How to actually work on it

Concretely, start with the floor, not the ceiling.

Stress first, because nothing upstream works on a fried nervous system. The fastest lever is breathing that lengthens the exhale — a few rounds where the out-breath runs longer than the in-breath. It nudges the body out of alert in under two minutes, and it's free. Protect sleep like it's a line item, because it is.

Attention next. Pick one thing to deliberately notice for a week — evidence that a relationship is working, or moments your body actually feels good. You're not lying to yourself. You're correcting a filter that's been quietly skewed toward threat.

Story next. When something stings, write the event in one line, then write the meaning you gave it in the next line. Separating the two is most of the work. The meaning is usually a guess wearing a fact's clothes.

Habits last, and smallest. Don't install five. Install one, attached to something you already do. After coffee, two minutes of breathing. The cue does the remembering so you don't have to.

None of this is dramatic. It compounds. A nervous system that resets, an attention that's aimed well, a story that's accurate, a habit that runs clean — that's a different person inside the same circumstances.

The obvious objection

"Isn't this just self-help with new vocabulary?" Fair question. The difference is where it points.

Most self-help is about getting more out of the human — more hustle, more morning routine, more squeezing. This is the opposite. The argument is that the squeezing era is ending because the machines now do the squeezable part. So the human's job changes. You stop optimizing yourself like a slightly slow computer and start tending the things a computer will never have.

The second objection I take seriously: if the machines do the work, does the human go soft and idle? I don't think so, and the reason is in the mechanism above. A regulated nervous system, clear attention, an accurate inner story — these don't make you passive. They make you capable of the harder human things. Judgment under uncertainty. Being fully present with another person. Taste. The decision no model can own because it has no stake in the outcome.

The machine handles the doing. That frees the human to do the becoming — which was always the harder and more valuable job, and the one we used to be too busy to get to.

What the human is actually for

So, the whole thesis. The machines got good at the work, and that's not a loss. It's a return of attention to the right place.

For most of history we measured a person by output, because output was scarce and expensive. It isn't anymore. When the doing gets cheap, the value moves to the things that can't be automated: judgment, presence, the slow work of changing who you are. Those things live in the system underneath a life — attention, stress, story, habits — and that system is trainable. That's the part I find hopeful. It means the most valuable capacities aren't a gift you're born with or without. They're tended, like a garden, by anyone who decides to.

That's the project behind MARSA, and behind everything I write at marsa.ai. Hand the work to the machines. Tend the human underneath, on purpose, with mechanisms that actually work rather than slogans that feel good for a weekend.

The question was never whether the machines would take the work. They will. The question is what you'll do with the part of your life they hand back. Thank you for being early.