Originally published on LinkedIn on July 27, 2026.
"The economic problem of society is not merely a problem of how to allocate given resources. It is a problem of the utilization of knowledge which is not given to anyone in its totality."
— Friedrich Hayek, 1945
The first article in this series proposed a question that management science has largely avoided: what makes an organization intelligent?
Before attempting to answer it, I want to examine something more fundamental. Why do organizations need collective intelligence at all?
The answer begins long before business schools or management theory. It begins with one of the oldest trade-offs in nature.
The evolutionary wager
The earliest forms of life were, in a sense, complete.
A single-cell organism performed every function necessary for its survival — sensing its environment, acquiring energy, reproducing, adapting. There was no coordination problem because there was only one decision-maker. What it lost in raw capability, it gained in self-sufficiency.
As life evolved, cells began to specialize. Some became muscle. Others became neurons, immune cells, liver tissue. No individual cell became more capable than before; each, in fact, became less capable of surviving independently. And yet the organism as a whole became vastly more powerful.
Because specialization unlocked complexity. But it also created something entirely new: dependency. The moment cells specialized, they needed each other. Communication became essential. Coordination became indispensable. Intelligence emerged not from any individual cell, but from the integration of millions of specialized ones working together.
The brain makes this vivid. A single neuron is extraordinarily specialized, capable of almost nothing in isolation. Intelligence emerges only because billions of neurons continuously exchange information through an intricate network of connections. Remove the network, and you don't have a slower brain. You have no brain at all.
Specialization doesn't just create capability. It creates an irreversible need for integration.
Organizations inherit the same problem
A sole proprietor resembles that single-cell organism. One person understands the customers, builds the product, manages the finances, and makes every decision. Knowledge is concentrated in one mind. Coordination is nearly unnecessary.
As organizations grow, they inevitably specialize. Engineering develops deep technical expertise. Sales understands customers. Finance understands capital allocation. Operations understands execution. Each function becomes dramatically better at its own discipline.
But something else happens simultaneously: no one understands the entire organization anymore.
This is a structural consequence of specialization, and Friedrich Hayek understood it with unusual clarity. Writing in 1945, he argued that the fundamental challenge facing any complex system is the utilization of knowledge dispersed across thousands of minds — knowledge that cannot be concentrated in any single place without losing most of what makes it valuable. No CEO, no executive committee, no planning function possesses the full picture. The knowledge required to make good decisions is always, irreducibly, distributed.
The challenge facing the organization is therefore not finding smarter individuals, but integrating what specialized ones already know into coherent collective decisions. This, I believe, is the moment Organizational Co-Intelligence becomes necessary — not as an aspiration, but as a structural requirement that specialization itself creates.
Intelligence as an emergent property
This distinction changes how we think about management fundamentally.
Organizations do not become intelligent by hiring brilliant people, adopting AI, or implementing the latest management framework. All of these can improve performance. But intelligence is a different kind of problem. It is an emergent property that arises when specialized knowledge is integrated into coherent collective judgment — and emergence cannot be purchased or installed.
Just as consciousness does not reside in a single neuron, organizational intelligence does not reside in the CEO, the executive team, or any department. It emerges from the quality of connections between specialized parts, and those connections are far harder to build than the specialized capabilities themselves.
This also reframes what leadership actually is. Leadership's role here is more modest than we often assume: it is one of the mechanisms through which organizations strengthen or weaken their collective capacity for thought, not a substitute for it. Great leaders rarely possess all the answers. They create the conditions under which distributed knowledge can surface, connect, and become better decisions.
Those conditions, based on what this research has found so far, cluster around four recurring patterns. Intelligent organizations share a common purpose that aligns decentralized decisions without requiring centralized control. They distribute knowledge intentionally, ensuring that expertise developed in one part of the organization can influence decisions across it. They maintain information flows that cross functional boundaries rather than pooling inside silos. And they learn — genuinely learn — from feedback, rather than repeating yesterday's decisions with greater efficiency.
Every one of these is an organizational architecture problem. Technology can support them, but it cannot substitute for them — and that matters enormously for what comes next.
AI amplifies what already exists
Artificial intelligence has dramatically accelerated individual capability — writing, analysis, design, decision support. The temptation is to assume this solves the organizational intelligence problem by making each individual node in the network smarter.
That assumption misses something fundamental. Organizations are networks of specialized knowledge, and making each node smarter does not automatically make the network more intelligent. The fundamental challenge is no longer helping one person think faster. It is helping thousands of people think better together.
Here is what makes this moment genuinely consequential: AI amplifies what already exists. It magnifies whatever architecture is already in place — the capabilities and the dysfunctions, with equal fidelity.
An organization without a clear mission will use AI to move faster in multiple directions at once. An organization with a poor culture will use AI to scale its worst behaviors. An organization that never developed genuine specialization will feed AI shallow inputs and receive shallow outputs in return. An organization without coordination mechanisms will use AI to generate more activity, not better decisions.
The real question is not whether your organization is deploying AI. It is whether you have built something worth amplifying.
A different question
Perhaps management has been solving the wrong problem for decades.
Rather than asking how organizations become more efficient — better at executing what they already know — we might ask how they become more capable of thinking: how they take the dispersed knowledge that specialization inevitably creates and integrate it into something coherent enough to act on.
In biology, specialization without integration doesn't produce a more efficient organism — it produces pathology. Organizations are more forgiving, but not infinitely so. The larger and more specialized they become, the more their future depends on their capacity to connect what they know into decisions that are actually right.
That capacity is what I mean by Organizational Co-Intelligence. And in the age of AI, it may be the only capability that compounds over time rather than depreciates.
There is a certain irony worth sitting with. The very pressures that make organizational intelligence necessary — complexity, specialization, scale — are the same pressures most likely to erode it. As organizations grow more capable, they tend to grow more fragile in precisely the ways that matter most: harder to align, slower to learn, more prone to mistaking activity for thought.
The architecture that holds this tension together is less visible than an org chart and less legible than a balance sheet. But it may be more consequential than either.
What that architecture actually consists of — and why it tends to fail quietly, long before anyone notices — is where this investigation goes next
Explore the OCI series
Previous: The AI Revolution Is Asking the Wrong First Question
Discuss organizational co-intelligence in your organization.
Originally published on LinkedIn on July 27, 2026.
"The economic problem of society is not merely a problem of how to allocate given resources. It is a problem of the utilization of knowledge which is not given to anyone in its totality."
— Friedrich Hayek, 1945
The first article in this series proposed a question that management science has largely avoided: what makes an organization intelligent?
Before attempting to answer it, I want to examine something more fundamental. Why do organizations need collective intelligence at all?
The answer begins long before business schools or management theory. It begins with one of the oldest trade-offs in nature.
The evolutionary wager
The earliest forms of life were, in a sense, complete.
A single-cell organism performed every function necessary for its survival — sensing its environment, acquiring energy, reproducing, adapting. There was no coordination problem because there was only one decision-maker. What it lost in raw capability, it gained in self-sufficiency.
As life evolved, cells began to specialize. Some became muscle. Others became neurons, immune cells, liver tissue. No individual cell became more capable than before; each, in fact, became less capable of surviving independently. And yet the organism as a whole became vastly more powerful.
Because specialization unlocked complexity. But it also created something entirely new: dependency. The moment cells specialized, they needed each other. Communication became essential. Coordination became indispensable. Intelligence emerged not from any individual cell, but from the integration of millions of specialized ones working together.
The brain makes this vivid. A single neuron is extraordinarily specialized, capable of almost nothing in isolation. Intelligence emerges only because billions of neurons continuously exchange information through an intricate network of connections. Remove the network, and you don't have a slower brain. You have no brain at all.
Specialization doesn't just create capability. It creates an irreversible need for integration.
Organizations inherit the same problem
A sole proprietor resembles that single-cell organism. One person understands the customers, builds the product, manages the finances, and makes every decision. Knowledge is concentrated in one mind. Coordination is nearly unnecessary.
As organizations grow, they inevitably specialize. Engineering develops deep technical expertise. Sales understands customers. Finance understands capital allocation. Operations understands execution. Each function becomes dramatically better at its own discipline.
But something else happens simultaneously: no one understands the entire organization anymore.
This is a structural consequence of specialization, and Friedrich Hayek understood it with unusual clarity. Writing in 1945, he argued that the fundamental challenge facing any complex system is the utilization of knowledge dispersed across thousands of minds — knowledge that cannot be concentrated in any single place without losing most of what makes it valuable. No CEO, no executive committee, no planning function possesses the full picture. The knowledge required to make good decisions is always, irreducibly, distributed.
The challenge facing the organization is therefore not finding smarter individuals, but integrating what specialized ones already know into coherent collective decisions. This, I believe, is the moment Organizational Co-Intelligence becomes necessary — not as an aspiration, but as a structural requirement that specialization itself creates.
Intelligence as an emergent property
This distinction changes how we think about management fundamentally.
Organizations do not become intelligent by hiring brilliant people, adopting AI, or implementing the latest management framework. All of these can improve performance. But intelligence is a different kind of problem. It is an emergent property that arises when specialized knowledge is integrated into coherent collective judgment — and emergence cannot be purchased or installed.
Just as consciousness does not reside in a single neuron, organizational intelligence does not reside in the CEO, the executive team, or any department. It emerges from the quality of connections between specialized parts, and those connections are far harder to build than the specialized capabilities themselves.
This also reframes what leadership actually is. Leadership's role here is more modest than we often assume: it is one of the mechanisms through which organizations strengthen or weaken their collective capacity for thought, not a substitute for it. Great leaders rarely possess all the answers. They create the conditions under which distributed knowledge can surface, connect, and become better decisions.
Those conditions, based on what this research has found so far, cluster around four recurring patterns. Intelligent organizations share a common purpose that aligns decentralized decisions without requiring centralized control. They distribute knowledge intentionally, ensuring that expertise developed in one part of the organization can influence decisions across it. They maintain information flows that cross functional boundaries rather than pooling inside silos. And they learn — genuinely learn — from feedback, rather than repeating yesterday's decisions with greater efficiency.
Every one of these is an organizational architecture problem. Technology can support them, but it cannot substitute for them — and that matters enormously for what comes next.
AI amplifies what already exists
Artificial intelligence has dramatically accelerated individual capability — writing, analysis, design, decision support. The temptation is to assume this solves the organizational intelligence problem by making each individual node in the network smarter.
That assumption misses something fundamental. Organizations are networks of specialized knowledge, and making each node smarter does not automatically make the network more intelligent. The fundamental challenge is no longer helping one person think faster. It is helping thousands of people think better together.
Here is what makes this moment genuinely consequential: AI amplifies what already exists. It magnifies whatever architecture is already in place — the capabilities and the dysfunctions, with equal fidelity.
An organization without a clear mission will use AI to move faster in multiple directions at once. An organization with a poor culture will use AI to scale its worst behaviors. An organization that never developed genuine specialization will feed AI shallow inputs and receive shallow outputs in return. An organization without coordination mechanisms will use AI to generate more activity, not better decisions.
The real question is not whether your organization is deploying AI. It is whether you have built something worth amplifying.
A different question
Perhaps management has been solving the wrong problem for decades.
Rather than asking how organizations become more efficient — better at executing what they already know — we might ask how they become more capable of thinking: how they take the dispersed knowledge that specialization inevitably creates and integrate it into something coherent enough to act on.
In biology, specialization without integration doesn't produce a more efficient organism — it produces pathology. Organizations are more forgiving, but not infinitely so. The larger and more specialized they become, the more their future depends on their capacity to connect what they know into decisions that are actually right.
That capacity is what I mean by Organizational Co-Intelligence. And in the age of AI, it may be the only capability that compounds over time rather than depreciates.
There is a certain irony worth sitting with. The very pressures that make organizational intelligence necessary — complexity, specialization, scale — are the same pressures most likely to erode it. As organizations grow more capable, they tend to grow more fragile in precisely the ways that matter most: harder to align, slower to learn, more prone to mistaking activity for thought.
The architecture that holds this tension together is less visible than an org chart and less legible than a balance sheet. But it may be more consequential than either.
What that architecture actually consists of — and why it tends to fail quietly, long before anyone notices — is where this investigation goes next
Explore the OCI series
Previous: The AI Revolution Is Asking the Wrong First Question
Discuss organizational co-intelligence in your organization.