Games of uncertainty

26/7/2025 ☼ games ☼ not-knowing ☼ uncertainty ☼ risk

tl;dr: I’m looking for — or would like to build — games that operationalise true uncertainty, not just simple risk, to help players become better at making decisions when facing different types of not-knowing.

Games are safe spaces to learn how to deal with the unknown. By design, they can be insulated from real-world consequences, giving us room to experiment, fail, and try again. They help us practice for reality, which is often has surprises, shocks, and unexpected events.

But there’s a problem with games as a way to learn to deal with unknowns: They nearly always operationalise only risk. And risk is only one of many types of unknowns.

Risk is known unknowns

In a situation of risk, four critical things are all known in advance.

We know every possible action we could take.
We know every possible outcome that might follow from those actions.
We know the precise and accurate probabilities connecting each action to each outcome.
We know the value of each outcome; whether it’s morally good or bad, whether we like it or not. (“Value” isn’t just about utility or money; it also covers qualitative preferences, ethical stances, or strategic priorities.)

In a game context, these conditions come from clear rules of play: What moves a player can make, what events might happen, what winning looks like, and what counts as a better or worse result.

Think about chess, Uno, Settlers of Catan, Angry Birds, Risk, or go. All these are games of risk. The pieces, the moves, the scoring, the win conditions; these are all known in advance and they don’t change. Players don’t know how each game will turn out — but they know all the possible outcomes and how likely they are, given the other players’ choices. (This is also true of rule-based sports like basketball, baseball, etc.)

Do these games of risk make us better at analysing risk? Possibly. At least, they train us to weigh known probabilities and preferences, to plan our actions based on those probabilities, and to respond when our calculations don’t pan out.

The important unknowns are uncertain, not just risky

Most of the really consequential unknowns we face in the world aren’t risky — they’re genuinely uncertain.

In true uncertainty, at least one of those four key pieces is missing. We might not know what actions are possible. We might not know what outcomes could result. We might not know the probabilities that connect actions and outcomes either accurately or precisely. Or we might not know how to value the outcomes — what is morally good or bad, or what we should prefer.

(I’ve written a detailed piece explaining the difference between risk and uncertainty here.)

Increasingly, the important unknowns we face today are uncertain, not just risky. Unquantifiable uncertainty, not just quantifiable risks suffuse both the causes and consequences of acute and contagious economic crises (e.g., the global financial crisis of 2008) and pandemics (e.g., Covid-19 in 2020), weakenings of the established rules-based world order (e.g., the Russian invasion of Ukraine in 2022), transformational technologies (e.g., ChatGPT in 2023), catastrophically extreme weather (e.g., the floods in Spain in 2024, or the LA wildfires in 2025).

For these unknowns, we could not know in advance what the full set of outcomes could be, what actions were truly available to us, what probabilities connected our actions to outcomes achieves, or even what outcomes we should value most.

The four types of not-knowing

To make sense of true uncertainty, it helps to break it down further into four distinct types of not-knowing:

Not knowing about possible actions, where we don’t yet know what moves can be made.
Not knowing about possible outcomes, where we don’t know what might happen as a result of our actions.
Not knowing about causation, where our knowledge of the mechanisms or probabilities connecting actions to outcomes is incomplete, inaccurate, or imprecise.
Not knowing about value, where we don’t know which outcomes are relatively more valuable and whether that system of valuation is stable and agreed-upon.

(I’ve written more about each type of not-knowing here.)

These types of not-knowing often show up together, but each type is conceptually different — and each demands a different kind of response. Being able to distinguish between them matters. If we can’t tell risk apart from uncertainty, and can’t separate these four types of not-knowing, we end up using the wrong tools to decide what actions to take.

What goes wrong when we treat uncertainty as risk

One of the most damaging mistakes is to treat an uncertain situation as if it were merely risky — applying risk-management methods that assume known actions, outcomes, probabilities, and values.

For instance, in February 2020, the World Health Organization used cost-benefit analyses to decide to recommend against countries imposing even temporary travel or trade restrictions to slow the spread of Covid-19. At that time, the basic epidemiological facts about the virus weren’t yet known. Using cost-benefit analysis made sense only if probabilities and values were known—which they weren’t. This is an example of what I’ve described elsewhere as the harms of misinterpreting uncertainty as risk.

A cost-benefit analysis used to make decisions in an uncertain situation where neither costs nor benefits can be precisely estimated.

I’m not saying that risk-analysis and risk-management are useless; they essential decision-making tools when the situation really is risky. But applying risk methods to uncertain situations can lead to actively harmful decisions.

How games of true uncertainty could help

Games have a special potential to help here. Games that operationalise true uncertainty — forcing players to navigate situations where actions, outcomes, probabilities, or values are unknown — could help players build an intuitive sense of what uncertainty feels like. Players could practice noticing which type of not-knowing they’re facing and experiment with different non-risk strategies to respond.

Even better would be games that clearly distinguish between risk and uncertainty, and between the four types of not-knowing. Such games wouldn’t just make uncertainty feel real; they’d also help train the cognitive skill of recognising which kind of unknown is present — and what kind of response it requires.

What we already have (and what’s missing)

I haven’t found many games that do this. A handful do seem to operationalise true uncertainty in game design:

These games don’t just throw randomness at players — they create space for unknown actions, outcomes, causation, and even values to emerge during play. They force players to deal with not-knowing that can’t be reduced to probability tables.

But even these games fall short in two ways:

They don’t always make players consciously distinguish uncertainty from risk.
They don’t separate out the four types of not-knowing, so players can learn to respond differently to each.

Where we need to go next

Games often get described as tools for learning to handle the unknown. But if we look closely, most games only teach us how to manage risk — where all the relevant rules, actions, and definitions of success are already known in advance, even if precise outcomes aren’t.

Uncertainty is different. In uncertainty, some of those aspects of the game simply aren’t known and may be unknowable until later (or ever). And uncertainty itself isn’t monolithic. It consists of different types of not-knowing: about actions, outcomes, probabilities, and values.

We have very few games that truly capture uncertainty, and none (as far as I know) that push players to distinguish the four types of not-knowing as separate decision-making challenges.

We need more games that:

Operationalise uncertainty, not just risk.
Help players feel the difference between risk and uncertainty in practice.
Separate out and make players grapple with the four distinct types of not-knowing.

Such games would be truer to life — and they’d help us build deeper, more flexible decision-making skills. Skills that matter far beyond the game table: in strategy, in policy, and in life itself.

For the last few years, I’ve been wrestling with the practical challenges of meaning-making in our increasingly AI-saturated world, developing frameworks for how humans can work effectively alongside these powerful tools while preserving the meaning-making work that is the irreplaceably human part of the reasoning we do. I’ve published this as a short series of essays on meaning-making as a valuable but overlooked lens for understanding and using AI tools

I’ve also been working on turning discomfort into something productive. idk is the first of these tools for productive discomfort.

And I’ve spent the last 15 years investigating how organisations can succeed in uncertain times. The Uncertainty Mindset is my book about how to design organisations that thrive in uncertainty and can clearly distinguish it from risk.