Glossary
Digital DNA
The complete working model of an operator. Its defining feature is not what it holds. It is that two of the things it holds are stored apart, and never allowed to mix.
Digital DNA (noun). A model of a specific person covering four things: what they produced, how they decide, who is doing the deciding, and the future they have chosen. The Imprint supplies the second. Belief propagation keeps the whole thing current. The chosen future gives it direction.
The last two are held in separate files, deliberately, because a system that cannot distinguish them will treat history as destiny.
The reason for the name
One property, taken literally. DNA is code you do not observe. You observe what it builds.
The patterns underneath an operating career work the same way. Nobody watches their own defaults run. What they see is the output: the business as it is, the calendar as it fills, the standard that gets held and the one that quietly does not, the project that stalled at the exact moment it started working.
Those outcomes usually get filed under results, timing, or the market. Some of that is true. Underneath it, something consistent was executing, and it was consistent enough that a colleague of ten years could probably describe it to you faster than you could describe it to them.
What changed is that the code can now be made readable. Which means it can be corrected, kept current, and, more to the point, aimed.
Two files, and the reason they stay apart
Inside a working system, who you are now and who you have chosen to become live in separate places. That is an architectural decision, not a filing convention, and it is the part of this term worth taking away.
| File one | File two | |
|---|---|---|
| Holds | The truth about you now | You at your best, chosen |
| Contents | Strengths, faults, scripts, defaults, how you actually behave | A daily operating standard, confirmed by you |
| Source | Evidence. Observed behaviour and captured rulings | Decision. What you have committed to becoming |
| Used for | Predicting what you will do | Deciding what to reinforce |
| If it is missing | The system flatters you | The system optimises your history |
Blur the two and a specific failure follows. An intention written down last month starts being read as an established pattern, so the system stops noticing the gap between them. Or the reverse: a default you are actively working to break gets treated as a fixed characteristic, and the system helpfully arranges your work around it.
Kept apart, the system can hold a live decision up against both and ask a question that has no other way of being asked.
And which one does it strengthen?
The default failure, which is comfortable
Most AI compounds the first file, because the first file is the one with evidence behind it. That is a reasonable engineering decision and it has an unpleasant consequence.
It takes you exactly as it finds you. Including the goals that got smaller after something went badly. Including the targets that are described as realistic because they were calibrated against what you have already done. Including the standard you dropped during a rough quarter and never picked back up.
Then it makes you faster at all of it.
This is why a system that only holds the first file feels excellent and is quietly limiting. There is no error state. The work improves. The direction is whatever your history was already pointing at, and it will keep pointing there with increasing efficiency.
What the second file has to be to work
The failure mode on the aspirational side is writing something inspiring and useless. A statement of values that cannot be applied to a decision does nothing, because decisions are where the model actually gets consulted.
The version that works is behavioural and specific. Not what you want to achieve. How the version of you at your best handles the thing in front of you right now: what they take on, what they decline, what standard they hold when holding it is expensive, where they stop deliberating and move.
And it has to be confirmed by the person, not derived. A chosen future that was inferred from your existing trajectory is just file one wearing better clothes, which returns the whole system to the default failure with an extra step.
Frequently asked
What is Digital DNA?
The complete working model of an operator: what they produced, how they decide, who is deciding, and the future they have chosen. It is not a separate product from an Imprint. The Imprint holds how you decide; the wider model holds who is doing the deciding and what they are aiming at.
Why call it DNA?
Because of one property, not as a general metaphor. DNA is code you never see directly. What you see is what it built. The same is true of the patterns underneath a career: you do not observe them, you observe the business, the calendar, and the project that stalled the moment it started working. Those are expressions. The code was running the whole time.
Why keep who I am and who I want to become in separate files?
So the system can tell evidence from intention. Blur them and it cannot: an aspiration you wrote down last month starts getting treated as an established pattern, or a pattern you are trying to break gets treated as your destiny. Kept apart, the system can look at a decision and ask which of the two it came from, and which one it strengthens.
Isn't the second file just a vision board?
A vision board describes outcomes. This describes an operating standard: how the person at their best actually behaves when a decision is in front of them, confirmed by the person rather than aspirational. The difference is that a standard can be applied to a live decision. A picture of a result cannot.
What does most AI do by default?
It compounds the first file, because the first file is the one with evidence behind it. So it takes you as it finds you, including the ambitions that shrank after something went badly, and then it makes you faster at all of it. Arriving sooner at the wrong place is a real risk and it does not feel like one on the way.
Where this sits
Digital DNA is the widest of the terms in this glossary, and the other entries are pieces of it. The Imprint is how you decide. The Five Levels of Knowing is the map of what each piece can and cannot be filled from. The category itself is defined at What is Imprinted AI.
Related terms
- The Five Levels of Knowing. The two files are the top two levels, named separately for this reason.
- Imprint. How you decide, which sits one layer below who is deciding.
- Belief propagation. What keeps both files current when something changes.
Where the term comes from
This is one entry in the vocabulary of a longer argument. The full glossary has fifteen terms. The case they belong to runs about 23,000 words, it is free, and there is no email gate on it.
Read The A.I. Business Manifesto
Nothing on this page is for sale. Quote it, argue with it, or pass it on.
Last updated: 28 July 2026