After Taleb: Why I Built Operating in the Dark for a World That No Longer Returns to Normal

Nassim Nicholas Taleb gave us one of the most useful modern languages for thinking about uncertainty. Long before I began writing Operating in the Dark, his work had already established a territory that interested me professionally: randomness, fragility, exposure, the limits of prediction, the difference between what a system appears capable of surviving and what happens when an event arrives from outside its assumptions. Fooled by Randomness, The Black Swan and Antifragile changed the vocabulary around risk because they attacked an extremely persistent illusion — that the world becomes more controllable simply because our models become more sophisticated. I had no interest in rewriting that territory. My problem appeared somewhere else, after uncertainty had already entered the room.

In practical work, uncertainty rarely remains a single problem for long. A manager may lack information. Another may have enough information but no authority to act. A founder can possess both and still lack capital, time or organizational capacity. A company may have money, people and authority while its internal architecture prevents those resources from reaching the point where they are needed. Sometimes the organization works and the person carrying its critical decisions is already overloaded. In other situations the whole structure remains technically functional while everybody inside it is unconsciously waiting for an external environment that has ceased to exist. These conditions are often compressed into the same language of uncertainty, leadership or crisis. Operationally, they are radically different.

I began separating them.

That separation eventually produced six zones: uncertainty, responsibility, constraint, decision architecture, the operator, and permanent crisis. Inside each I began assembling frameworks around problems I had encountered repeatedly through years of managing projects, teams, products, transformations and complex systems. The purpose was not to create another universal management doctrine. I wanted to know where one type of problem ended and another began, because interventions become dangerous when those boundaries are ignored. A psychological problem requires one kind of work. An authority problem requires another. An information problem can look remarkably similar to incompetence from outside. A structural failure can survive several generations of talented managers because everybody keeps attempting to repair the people occupying the structure.

This became the six-book core of Operating in the Dark.

There is another reason I wanted to separate these zones. I believe the problem is considerably more relevant now than it was when Taleb’s major ideas began entering mainstream discussion around the turn of the century. Uncertainty itself is ancient. What has changed is the density of connections between unstable systems and the speed at which instability crosses from one domain into another. A geopolitical event becomes a logistics problem, an energy problem, an inflation problem and a corporate planning problem. A technological breakthrough changes investment behavior, labor expectations and competitive positioning before institutions have even developed stable terminology for it. Information travels globally in seconds while verification, legislation and organizational adaptation operate at completely different speeds. Companies depend on platforms, infrastructure, supply chains, payment systems, regulatory environments and political decisions they cannot control. AI is now accelerating another transformation while many organizations are still carrying structural debt from the previous one.

For an operator, this creates a peculiar environment. You are rarely dealing with one crisis. You are standing at the intersection of several systems moving at different speeds, with different levels of visibility and different relationships to your authority. The traditional managerial assumption that instability is an interruption between periods of normal operation becomes increasingly difficult to defend. In many industries and regions, instability is becoming part of the operating environment itself.

I wrote the Operating in the Dark books during the war in Ukraine. Air attacks, power outages and the practical consequences of a country at war were part of the environment in which the work was written. I mention this carefully because war should not be turned into decorative authenticity for a management argument. The underlying ideas did not originate with the invasion; they came from years of professional work with systems, responsibility and decisions. War simply removed a great deal of abstraction from them.

A deadline during a blackout remains a deadline, but everything surrounding it has changed. A person can disappear from a process because physical safety has become more important than the project. Infrastructure that yesterday felt permanent can become unavailable within minutes. International clients continue operating according to commercial calendars that have no relationship to your local conditions. Plans still have to exist, teams still have to communicate, financial obligations do not become philosophical concepts, and decisions continue accumulating consequences. Under those conditions, you discover quickly which elements of an operating model are genuinely structural and which survived only because the environment had previously been generous enough to support them.

The first zone of the series is therefore uncertainty, but I treat it as an operating condition rather than an intellectual mystery. The useful question is not how to know everything. It is how much information is enough to act, which unknowns genuinely matter, how reversible the decision remains, and when waiting for better information begins creating more risk than acting without it. This is the territory of the first book, Operating in the Dark: Decision Making Under Uncertainty, Pressure, and Limited Resources.

The second zone begins when uncertainty acquires an owner. The Weight of Responsibility examines what changes when consequences can be traced back to a particular person. Responsibility affects attention, delegation, risk perception and the way information moves through an organization. It can also create a destructive gravitational field: unresolved problems move upward, weak signals compete for senior attention, and capable leaders gradually become decision bottlenecks because everyone has learned that difficult questions eventually end with them. At that point responsibility is no longer a moral abstraction. It is part of system design.

Then comes constraint. Decision Under Constraint grew from a recurring professional irritation with strategies that are logically impressive and operationally impossible. Capital, authority, personnel, infrastructure, time, political legitimacy, organizational maturity and physical security define the real decision surface. If a strategy requires resources the system does not possess, its elegance is irrelevant. Constraints are not excuses added after execution fails. They belong inside the architecture from the beginning.

The fourth zone is where the perspective changes. Decision Architecture Without Relief asks what happens when the recurring failure belongs to the system producing the decisions. If several competent people keep failing in approximately the same place, replacing them can become a sophisticated method of avoiding the architecture. Information may arrive at the wrong level. Authority may be disconnected from execution. Reporting can produce enormous quantities of data without improving visibility. Processes designed for a smaller company can remain in place after growth has made them dysfunctional. Under pressure, architecture becomes visible because the organization loses the spare capacity that previously concealed its defects.

The fifth book, The Operator’s Mindset, puts the person back inside that architecture. I was interested in the operator without the mythology surrounding leadership. Attention is finite. Judgment degrades. Experience improves pattern recognition and sometimes makes unfamiliar patterns harder to recognize. Previous success can become a constraint because identity begins defending the conditions that produced it. Discipline, in this context, is practical: filtering information, protecting decision bandwidth, sequencing actions, separating discomfort from evidence and preventing every urgent signal from becoming equally important. The operator is part of the system and can become one of its dependencies.

The sixth zone was the logical endpoint: Crisis as a Permanent State. What happens if the assumption of eventual relief is removed? Many organizations can function under exceptional conditions for several months because everybody expects those conditions to end. Temporary authority is accepted. People absorb additional load. Structural problems are postponed. Reserves are consumed. The organization borrows from the future because the future is expected to be easier. The danger begins when the temporary environment becomes permanent while the operating architecture continues waiting for normality.

This is especially relevant to me now. Living and working through years of war makes the phrase “when things return to normal” sound increasingly like a strategic assumption that should be tested rather than a promise that should be trusted. The same logic extends far beyond Ukraine. Markets wait for normalization, companies wait for technological transitions to settle, political systems wait for polarization to decrease, industries wait for predictable regulation, and organizations wait for a stable moment in which they will finally have time to repair what they have been postponing. Sometimes that moment arrives. Building a system whose survival requires it to arrive is a different proposition.

The six books move through these zones in sequence because each changes the next. Uncertainty becomes more consequential when responsibility enters. Responsibility becomes more difficult when the available options are constrained. Repeated constraints reveal weaknesses in decision architecture. Architecture cannot be understood without examining the operator functioning inside it. And the entire structure changes again when crisis stops behaving like a temporary event.

After decomposing the problem across six books, I wanted to perform the reverse operation and assemble it again.

Books Seven and Eight exist for that reason. Operating in the Dark: The First Cycle brings together the first three books — uncertainty, responsibility and constraint — while adding chapters, connections and conclusions that become visible when the three are treated as one field. Operating in the Dark: The Complete Cycle does the same with the second half, connecting decision architecture, the operator and permanent instability, again with additional material and conclusions. They are not simply bundles of previously published books. The six original works provide depth inside individual zones; the two larger volumes restore the relationships between them. Decomposition made the mechanisms visible. Reassembly made the system visible.

That could have been the end of the architecture. Instead, I added a book that deliberately stands slightly outside it.

Responsibility: Anti-Coaching — Principles, Logic, Frameworks, and Applied Cases exists because frameworks eventually have to survive contact with dirty reality. It takes the six central zones and moves them into applied situations drawn from professional practice: business operating during war, international expansion outrunning organizational capacity, institutional information and narrative systems, humanitarian operations, corporate deterioration and high-performance sport. Here the variables refuse to remain separated. People become tired. Information arrives late. Budgets change. Institutions defend previous decisions. Authority becomes ambiguous. An intervention solves one problem and produces another.

Anti-Coaching also addresses something that had become increasingly obvious to me while working with people and organizations: many problems presented as personal failures are architectural failures wearing a human face. Someone is told to become more confident when they lack actual authority. A team is asked to become more motivated while operating with insufficient resources. A manager is taught resilience while the organization continues transferring every unresolved decision upward. A founder is encouraged to think bigger when the existing structure is already failing under its current scale. Advice becomes dangerous when it increases conviction without increasing the system’s capacity to support action.

My Anti-Coaching approach therefore starts with diagnosis rather than encouragement. What is actually controllable? Where does authority sit? Which constraints are material? Which commitments are reversible? Where is information distorted? What depends on what? Is the person failing to operate the system, or is the system manufacturing the failure attributed to the person? Only after those questions does intervention make sense. Sometimes the person needs to change. Sometimes the architecture does. Sometimes the objective is impossible under the available constraints. Occasionally the most responsible decision is to stop pretending that another motivational push will alter any of these facts.

That is why Responsibility: Anti-Coaching belongs to the series and still stands apart from its main sequence. The first six books divide the territory. Books Seven and Eight reconstruct the larger architecture. Anti-Coaching pushes the architecture back into practice and asks whether it remains useful once reality contaminates the model.

I started this work from uncertainty. Nine books later, I think uncertainty may be the easiest part of the problem.

The harder question is what happens when uncertainty becomes infrastructure: when instability lasts longer than the plans built to survive it, when several external systems move simultaneously, when information increases faster than the ability to interpret it, and when the people responsible for decisions cannot postpone consequences until the world becomes easier to understand. At that point another prediction will not necessarily help. Another motivational narrative certainly will not. What matters is whether the architecture can still produce coherent action without borrowing its logic from an imagined return to normal.

I wrote Operating in the Dark in a country at war, but I do not think its central problem belongs only to countries at war. It belongs to any environment in which the old promise — endure the disruption and normality will return — has stopped being reliable enough to build around.

The dangerous moment is not when you realize that you do not know enough.

It is when you realize that you may never know enough — and the decision is still yours.