Sensibly Deal with Uncertainty and Risk

6. Civic Knowledge and Skills

This introductory article was written by ChatGPT at the direction of Heidi Burgess, who reviewed, edited, and approved the final content.
For more information

June 24, 2026

One of the most common misunderstandings about science is the idea that it should be able to make perfectly certain pronouncements. In reality, scientific findings almost always involve some degree of uncertainty. This does not mean that science is unreliable. It means that honest inquiry recognizes limits: data may be incomplete, measurements may be imprecise, models may simplify reality, and future conditions may not match past experience. In public conflict, uncertainty is often used badly. Some people exaggerate uncertainty to avoid action, while others hide uncertainty to make their preferred policy sound more settled than it is. Both approaches damage trust and lead to poor decisions.

It is useful to distinguish between risk and uncertainty. Risk refers to situations in which the possible outcomes are reasonably well understood and their probabilities can be estimated, even if not perfectly. For example, engineers may be able to estimate the chance that a bridge will fail under certain loads, or public-health researchers may estimate the risk that a disease will spread under different conditions. Life insurance agents can estimate the risk of someone dying over the policy period, just as home insurance agents can estimate risks of fires and theft.

Uncertainty is deeper. It refers to situations in which important outcomes, probabilities, causal relationships, or future conditions are not known well enough to calculate a reliable risk. Economist Frank Knight’s classic distinction between risk and uncertainty is often summarized this way: risk involves measurable probabilities, while uncertainty involves probabilities that cannot be measured with confidence.

Risks are best handled through disciplined assessment and management. This means identifying possible harms, estimating their likelihood and severity, comparing options, and deciding which risks are acceptable in light of the benefits and costs. Risk management may involve prevention, mitigation, insurance, safety margins, emergency planning, or choosing a less dangerous alternative. The National Academies' work on risk assessment emphasizes that such analysis is especially important when decisions must be made with limited resources to protect public health, safety, and the environment. Risk analysis does not remove values from the decision. It clarifies what is at stake so citizens and leaders can make those value choices more responsibly.

Uncertainty requires a somewhat different approach. When people do not know enough to estimate probabilities confidently, they should not pretend that they do. Instead, they should ask what range of futures might occur, what early warning signs should be watched, what actions would be useful under many different scenarios, and which choices would be hardest to reverse if they turned out badly. This is where strategies such as scenario planning, adaptive management, pilot projects, monitoring, contingency planning, and "no-regrets" actions become important. BI's discussion of fact-finding amid irreducible uncertainties makes this point in conflict terms: some uncertainty can be reduced by better information, but some must be lived with and managed.

In polarized conflicts, uncertainty is often weaponized. One side may say, "The science is uncertain, so we should do nothing." Another may say, "The science is settled, so no one has a right to object." Both responses are too simple. A more constructive approach asks several separate questions: What do we know with high confidence? What is plausible, but uncertain? What would be the consequences of being wrong in each direction? Who bears the risks? Who gets the benefits? And how can we adjust course as we learn more? These questions help shift the conflict from a fight over absolute certainty to a more honest discussion of evidence, values, tradeoffs, and responsibility.

Sensibly dealing with risk and uncertainty is essential because decisions cannot wait until knowledge is perfect. Choosing not to act is itself a decision, and it carries risks of its own. Good decision making therefore requires humility without paralysis and confidence without overstatement. Experts should communicate uncertainty clearly, policymakers should explain how they are weighing risks, and citizens should resist the temptation to treat uncertainty as either proof of ignorance or proof of bad faith. In complex conflicts, the goal is not to eliminate uncertainty. The goal is to make wise, revisable decisions while remaining honest about what is known, what is not known, and what could go wrong. (Our friend, Sanda Kaufman, talks about the value of "pre-mortems" — looking at what might go wrong before it does, and figuring out how risks can be limited and mistakes avoided, instead of "post-mortems" — figuring out what went wrong after the fact.

—————————

This page was created by ChatGPT in response to this prompt. It was then reviewed, edited, supplemented and approved by Heidi Burgess. More information about how and why we are using AI in this way, and about the growing number of ways in which Beyond Intractability is using ChatGPT, Claude and other AI systems to generate content and build out the BI system, is available on our BI/AI Overview Page.