- The Great Rethink
- Posts
- Local Fluidity · Global Coherence
Local Fluidity · Global Coherence
Why Linear Focus Is the Wrong Metric for the Age of AI
The Assumption
Focus means: stare at one thing. Move in a straight line. Maintain visible consistency. Reduce the aperture until only the task remains.
This is the industrial model of intelligence. It was designed for assembly lines, quarterly plans, and stable systems where the map matched the territory. In that world, focus was the primary virtue because the system rewarded repetition and the cost of distraction was waste.
Everyone still believes this. Boards tell founders to focus. Advisors tell operators to narrow. Investors ask for the one thing. The entire performance culture of business is built on the assumption that intelligence concentrates — that the best thinkers aim at a single point and refuse to look away.
But there is a problem with this assumption. It only works when the system you are operating inside is stable.
When the map is accurate, focus on the path. When the map itself is shifting, focus on the path and you will walk off a cliff with excellent discipline.
What Focus Actually Is
Here is the inversion most people miss.
Focus, in the traditional sense, is a local operation. It narrows the field of vision to one point, one task, one problem. It optimises within a known system. It is extremely effective when the system is stable and the variables are understood. This is why it works on assembly lines. The system does not change. The task repeats. Narrow attention produces narrow excellence.
But when the system itself is changing — when categories are emerging, when AI is shifting the substrate, when the rules that governed last quarter no longer apply this quarter — local focus becomes a liability.
Not because the person is undisciplined. Because they are disciplined about the wrong unit.
The person who is “focused” in the traditional sense, staring hard at a single variable inside a system that has already shifted, is not exhibiting discipline. They are exhibiting a failure to update. They are optimising a position on a map that no longer corresponds to the terrain. The harder they focus, the more precisely they navigate toward a destination that has moved.
This is the part that stings: the most focused person in the room may be the one who has lost the signal. Not because they lack intensity. Because they are applying intensity to a resolution that no longer matches the operating environment.
Local focus in a shifting system is not discipline. It is a high-resolution commitment to an obsolete frame.
How Complex Systems Actually Work
Nature does not focus locally. Consciousness does not. Complex adaptive systems do not. And AI certainly does not.
A river meanders. From any single bend, the movement looks random — a pointless curve in the wrong direction. From altitude, the river is finding the most efficient path to the ocean. Every meander is a local response to terrain that serves the global coherence of the system. The river does not “focus” on reaching the sea. It responds to what is in front of it while the gradient does the rest.
An ecosystem adapts. Species mutate, compete, cooperate, go extinct. From inside any single interaction, it looks chaotic. From the system level, the ecosystem is maintaining a coherence that no individual participant controls or even perceives. The coherence is not the result of any organism focusing. It is the result of a structure that holds while the parts move freely.
Evolution works the same way. Gravity works the same way. Markets work the same way. The global pattern holds. The local expression is fluid. The system does not maintain coherence through rigidity. It maintains coherence through responsive movement within a stable architecture.
The industrial model inverted this. It said: hold the parts rigid and the system will be coherent. That works in factories. It fails everywhere else.
Complex systems are not coherent because their parts are focused. They are coherent because their architecture holds while their parts move freely.
The Real Metric
If local focus is the wrong metric, what is the right one?
Local fluidity · Global coherence.
Locally, the movement looks nonlinear. The attention jumps. The investigation spirals. The curiosity wanders across domains. A single week touches category design, consciousness, AI architecture, pricing theory, and may be scripture. From the outside, it looks scattered. From the inside, it is one continuous investigation viewed from multiple altitudes.
Globally, the attractor never changes. The same themes keep appearing: coherence, signal and distortion, systems that hold under pressure, the difference between what people claim and what actually is, the structures underneath the stories. The center holds. The periphery explores. The exploration feeds the center. The center gives the exploration meaning.
This is not multitasking. Multitasking is local attention divided across multiple local tasks. This is something structurally different: global attention maintained across multiple expressions of the same investigation.
The person who operates this way is not unfocused. They are focused at a different resolution. They are holding the forest while moving between trees. They are maintaining the gradient while the river meanders.T
The question is not whether you are focused. The question is: at what resolution
Why This Matters Now
The industrial model of focus worked because the operating environment was stable enough to reward local optimisation. Quarterly plans held. Category boundaries held. Competitive positions held. A company could focus on one thing for three years and the system would not shift underneath it.
Three things broke this simultaneously.
AI compressed the cycle. Model releases every six months. Capability shifts every quarter. The substrate underneath every business is moving faster than any local focus can track. The company that focuses on optimising its current position is optimising a position that will not exist in eighteen months.
Categories are destabilising. The boundaries that defined markets for decades are dissolving. New categories emerge, merge, and collapse faster than the old playbooks can map them. Local focus on a single category position works until the category itself moves.
The signal environment saturated. Everyone is producing content, claims, positioning. The noise is overwhelming. Local focus — staring hard at your own message — does not help when the environment you are signalling into has changed its listening patterns. What helps is the ability to read the entire field and adapt without losing coherence.
In stable systems, the locally focused person wins. In shifting systems, the globally coherent person wins. The environment shifted. The metric did not update.
Two Forms of Intelligence
This produces a distinction that matters for anyone building in the current moment.
The first form of intelligence is convergent. It narrows. It selects one frame, one method, one domain, and optimises within it. It is extremely powerful inside stable systems. It produces specialists, operators, and executors who are excellent at the thing they were trained for.
The second form of intelligence is traversal. It moves across domains while maintaining structural coherence. It synthesises patterns from multiple fields. It recognises that the same architecture appears in consciousness, in markets, in ecosystems, in AI systems — and uses that recognition to redesign systems rather than optimise them.
Convergent intelligence builds within the frame. Traversal intelligence redesigns the frame.
The industrial age rewarded convergent intelligence almost exclusively. The AI age is shifting the reward structure. AI is better than humans at convergent tasks. It is not better than humans at traversal. The machine optimises locally with superhuman precision. It cannot hold the global coherence of a system it did not design.
This means the scarcest human capability is no longer the ability to focus on one thing for a long time. It is the ability to maintain coherence while moving through complexity — to hold the architecture while everything inside it shifts.
AI is convergent intelligence at scale. The human advantage is traversal. The person who can walk through the forest without forgetting where north is cannot be automated.
The Permission
There is a generation of founders and builders who have been told, repeatedly, that they lack focus. They jump between domains. They see connections others miss. They cannot explain their strategy in one sentence because the strategy operates at a resolution the one-sentence question cannot reach.
They have internalised this as a weakness. They believe the board member who says “you need more focus.” They believe the advisor who says “pick one thing.” They believe the culture that treats linear execution as the highest form of intelligence.
Some of them are genuinely scattered. No architecture, no attractor, just motion.
But some of them — the ones who keep returning to the same structural themes, who see the same pattern across different domains, whose “lack of focus” is actually a continuous investigation viewed from outside — those people are not unfocused. They are operating at a different resolution. The system they are building is not visible from inside the spreadsheet. It is visible from altitude.
The permission is this: if your global coherence is intact — if the center holds, if the attractor is stable, if the investigation is continuous even when the expression is diverse — then you are not scattered. You are traversing. The people telling you to focus are not wrong. They are focused at a resolution that cannot see what you are building.
The discipline is not in the narrowing. It is in maintaining the architecture while everything moves.
The Metric
The old metric: how long can you stare at one thing?
The new metric: how much complexity can you move through without losing coherence?
The future does not belong to the people who can stare hardest at a single tree.
It belongs to the people who can walk through the forest without forgetting where north is.
Local fluidity · Global coherence.
The river meanders. The gradient holds. The ocean is always north.