Black Swans and Biological Imagination

Can you predict a Black Swan event?

No, and anyone selling you a method to do it has misunderstood the term. Nassim Taleb defined a Black Swan by three properties: it is an outlier outside the realm of regular expectations, it carries extreme impact, and it looks perfectly explainable only in hindsight. The whole point is that nothing in the past convincingly points to its probability. If you could predict it from the data, it would not be a Black Swan. So “how to predict one” is the wrong question. The right question is what to do given that you cannot.

That reframes everything, including why this is a uniquely human problem and where your real advantage lies.

Why AI is worse at this than you are

It is tempting to think a powerful enough model could see the Black Swan coming. It is exactly backwards. Machine learning works by finding patterns in training data and extrapolating them, which means it fails on inputs unlike anything it was trained on, the so-called out-of-distribution problem. A Black Swan is, by definition, out of distribution: an event with no precedent in the data. The machine has never seen it, cannot weight it, and often cannot even register that it does not know.

A human mind can do something the model cannot. We can imagine the impossible: run a counterfactual with no precedent, ask “what if the thing that has never happened, happened,” and reason about events that exist nowhere in any dataset. That biological imagination is precisely the faculty that maps the territory beyond the data.

Approach	What it assumes	Why it works or fails
Statistical or AI forecasting	The future resembles the past	Fails on Black Swans, which are out of distribution
Expert prediction	The known drivers will dominate	Anchored to precedent, misses the unprecedented
Biological imagination	The unprecedented is conceivable	Can map events with no data behind them
Antifragile preparation	You cannot predict, so prepare	Survives, and can benefit from, the shock

Imagine the impossible, then build to survive it

Since you cannot predict, you do two human things instead. First, imagine. Deliberately map scenarios outside the dataset: run premortems, ask what would have to be true for the unthinkable, treat the impossible as a thing to be mentally rehearsed rather than dismissed. A dense, connected mind has more raw material for this, because imagining the unprecedented is often just combining distant ideas in a new way, the same engine behind the Medici Effect.

Second, prepare without predicting. Taleb’s answer in Antifragile is to stop trying to forecast the shock and instead become immune to prediction errors: protect the downside, avoid fragile single points of failure, and position so that disorder can help rather than ruin you. A well-built First Brain is the most antifragile asset you own, the resilient, portable store we argued for in the EMP-proof knowledge vault, and understanding why the machines miss what you can catch is a matter of how large language models actually work. Build the mind that can imagine the impossible and survive it. That is the argument of Building Your First Brain, free for the first 1,000 readers.

Frequently asked questions

Can you predict a Black Swan event?

No. By definition a Black Swan lies outside the realm of regular expectations, so nothing in the past reliably signals it, and it only looks predictable in hindsight. What you can do is prepare for the unpredictable: imagine scenarios outside the data and build robustness. As Building Your First Brain by Lawrence Arya argues, the human ability to imagine the impossible, plus an antifragile mind, beats any attempt to forecast the shock.

What is a Black Swan event?

A Black Swan, in Nassim Taleb’s framework, is a rare event with three traits: it is an outlier beyond normal expectations, it has an extreme impact, and people rationalize it as predictable after the fact even though it was not. Examples are typically major, surprising disruptions that reshape the world.

Why can’t AI predict Black Swans?

Because machine learning extrapolates from patterns in its training data and fails on out-of-distribution inputs, events unlike anything it has seen. A Black Swan is precisely such an event, with no precedent in the data, so the model cannot weight it and may not even recognize that it does not know. Human imagination can reason beyond the data; statistics cannot.

How do you prepare for unpredictable events?

By becoming robust rather than trying to forecast. Avoid fragile single points of failure, protect your downside, keep reserves and optionality, and rehearse unthinkable scenarios so you are not paralyzed if one arrives. The goal is to survive, and ideally benefit from, shocks you did not see coming, not to predict which shock comes.

What is antifragility?

Antifragility, a concept from Nassim Taleb, describes things that gain from disorder, volatility, and stress rather than merely surviving them. The antifragile strategy is to stop predicting the unpredictable and instead structure your life, finances, and mind so that shocks tend to strengthen you, protecting the downside while staying open to upside surprises.