Looking back on that time, I spent twelve days using AI to build an entire website, working on code, automation, and multilingual translation. After completing that stage, I sat staring at the screen for a long time.

Not because I was tired. A thought had suddenly surfaced: if all of this could be completed for three thousand, what was the judgment I had accumulated over twenty years actually worth?

This question made me uncomfortable. But uncomfortable questions are usually the right questions.

Negative Entropy Is a Metaphor Here, Not an Organization Technique

In Chapter 6 of What Is Life?, Schrödinger uses negative entropy to discuss how living organisms maintain a relatively low-entropy state through exchanges with their environment. In the note to the chapter, he adds that he would use free energy when writing for physicists. That is a physical discussion of life. I borrow the term to discuss organizing experience and constructing meaning; what follows is not a thermodynamic claim.

Organizing a desk, building a knowledge-management system, and taking notes can all help with work. AI can also speed up the organization and summarization of information, but making something look more orderly does not mean that physical negative entropy has been measured.

True humanistic negative entropy is something else: the ability to transform a fragmented experience into a meaningful story; to recognize what’s worth caring about among a pile of contradictory information; to make a decision you’re willing to bear the consequences of when facing a situation with no standard answer.

Models can also generate stories, compare reasons, and suggest directions. What matters to me is that, after seeing those suggestions, I still need to decide how I rank my values and bear the consequences of the choice.

What Theology Taught Me

I have fifteen years of theological training background. This experience is usually inconvenient to mention in tech circles, but it’s actually the underlying operating system for all my judgments.

For me, that theological training was not only about memorizing scripture. It was also about learning to face uncertainty within faith: some questions still had no answer, yet I still had to choose how to live.

This is very similar to entrepreneurship. You can never prove a business model will definitely work. You can only make judgments with limited information, then validate through action. The difference is that entrepreneurship validates the market; faith validates life.

AI can help compare options when goals and constraints are relatively clear. But decisions about marriage, starting a business, or leaving a stable job often require me to clarify the goals, costs, and responsibilities themselves. A tool can help me see the options, but it does not live the life that follows from my choice.

The first layer of humanistic negative entropy’s meaning lies here: enabling you to act even when you don’t know.

Why “Efficiency” is Actually a Trap

I spent ten years in the circular-economy industry, often looking at waste from another angle: could it become a resource for a different use? A new frame can reveal an opportunity, but quality, use, and cost still need to be checked. Changing the name alone does not complete a circular process.

The same logic applies to knowledge.

For me, the problem is sometimes not a lack of information, but that I have not yet worked out what it has to do with my life. AI can quickly organize concepts and related material about negative entropy, though the content still needs checking. But after it has been organized, what then? Which parts are worth carrying into my life?

When efficiency becomes the only measure, it can turn into a trap. If I focus only on consuming information quickly, I may miss what needs to be considered slowly: the warmth behind a poem, the struggle within a piece of history, or the lesson in a failure.

This is why I say humanistic negative entropy isn’t chicken soup. Chicken soup makes you feel good. Humanistic negative entropy makes you willing to face things that make you feel bad, then extract meaning from them.

Meaning is Not Found, It’s Created

During my collaboration with Claude, the biggest realization wasn’t technical. It was being forced to redefine “where my value lies.”

AI can help write code, translate, analyze data, and generate images. Some work that I once assumed required a team can now be partly started by one person using AI. Where, then, should I place more of my attention?

Judgment. Taste. The decision of where you choose to spend your time. The intuition that looks at a bunch of viable options and says “No, this direction is wrong.”

My judgment does not come from data alone. It also comes from the books I have read, the people I have loved, the pits I have fallen into, and the questions I have asked myself at three in the morning.

Meaning is not only found passively; it is also constructed through choice, action, and revision. I use humanistic negentropy as a metaphor for this process: organizing fragmented experiences into a direction whose consequences I am willing to bear, not altering the laws of thermodynamics.


Chaos is raw material.

The tools I use, and the way I refine experience into direction, also shape the person I become. AI is a useful tool and can suggest directions; choosing what to refine, whom it is for, and which consequences to bear remains my responsibility.

Connecting Understanding to Its Consequences

After organizing more information, I still want to ask how it relates to life.

Humanistic negentropy is a metaphor for constructing meaning, not a measurement of physical entropy.

AI can propose stories and reasons; I still need to decide how I rank values.

Efficiency is useful, and I also want to preserve time for slow reading, reflection, and revision.

Turn experience into direction, and bear the consequences of the choice.

Let tools help with understanding, while giving my own choices reasons.

Four-panel comic: After organizing more information, I still want to ask how it relates to life. Humanistic negentropy is a metaphor for constructing meaning, not a measurement of physical entropy. AI can propose stories and reasons; I still need to decide how I rank values. Efficiency is useful, and I also want to preserve time for slow reading, reflection, and revision. Turn experience into direction, and bear the consequences of the choice. Let tools help with understanding, while giving my own choices reasons.

Revised November 5, 2026. The physical concept draws on the linked Chapter 6 of What Is Life? and its concluding note. The site build, the question about cost, and the work experiences record my thinking at the time; humanistic negentropy is this essay’s metaphor.