Counterfactual Testing is testing model behavior on hypothetical alternate inputs. It sits in the Incentives dimension (INC) of the Human Behavior Taxonomy™ as element HBT-INC-0081, within the Risk family. The core principle: testing model behavior on hypothetical alternate inputs. In incentive terms, it matters because it changes the payoff people perceive before they choose — which means it can be designed for, or exploited.
Scientific Definition
Testing model behavior on hypothetical alternate inputs.
Plain-English Definition
Testing model behavior on hypothetical alternate inputs.
Feynman Explanation
What if the same applicant were male/female/named differently?
Core Principle
Testing model behavior on hypothetical alternate inputs.
Mechanisms
Pending editorial review.
Pending editorial review.
Pending editorial review.
Pending editorial review.
Pending editorial review.
Pending editorial review.
Pending editorial review.
Inputs (Triggers)
Pending editorial review.
Outputs (Behaviors)
Pending editorial review.
Behavioral Signature
What if the same applicant were male/female/named differently?
Examples
- Standard practice for fairness auditing.
- Bias detection.
Pending editorial review.
Original analysis from The Incentives Lab — how this element behaves inside real payoff structures.
Why this element matters to incentive design
When this element shows up in a diagnostic, the instinct is to train people out of it. Training rarely moves it. The mechanism underneath it operates in the Incentives dimension — what makes behavior more or less likely?. You can recognize it in the field by its signature: what if the same applicant were male/female/named differently?. Every element in the Incentives dimension changes the perceived payoff of an action before the action happens, which is exactly where incentive design has leverage.
How it gets exploited
Left undesigned, bias detection. It is amplified whenever bias detection. Inside organizations that shows up as bias detection. The pattern is the same one Goodhart's Law describes: the measurable proxy attracts the effort, and the purpose behind it quietly loses funding.
How the Lab designs around it
The redesign move is to run counterfactuals on consequential models continuously. Treat it as infrastructure. Once you can see it in your own system, most of the argument about culture resolves itself.
Famous Experiments
Pending editorial review.
Design Principles
- Run counterfactuals on consequential models continuously.
Measurement Approaches
Pending editorial review.
Evidence
Pending editorial review.
Pending editorial review.
The Perverse Incentive Lens™
How this behavior is exploited — and how to redesign around it.
- Run counterfactuals on consequential models continuously.
Pending editorial review.
Pending editorial review.
Interactive Mini Network
Click any neighbor to re-center the graph and follow the threads of connection.
Knowledge Graph Neighbors
Auto-linked to the rest of the Human Behavior Taxonomy by family, domain, dimension, and shared keywords.
When the agent acts, who's responsible?
Categorizing AI use cases by risk level.
Systematic skew in model behavior across groups.
Adding noise to data to protect individual privacy.
Quantitative measures of model behavior across groups.
Training models across devices without centralizing data.
Bypassing model safety constraints.
Individual speed gains hide collective quality decline.
Malicious instructions hidden in user input or retrieved content.
Adversarial testing of AI systems.
Foundational skills erode through AI offloading.
Concentration risk on a single AI provider.
Where Counterfactual Testing is cited in the corpus
Essays, field guides, and diagnostics from The Incentives Lab that apply this element.
- Field guideIncentives: definition, types, examples
The parent field guide for this element.
- ReferenceThe laws of incentives
Goodhart, Campbell, and the Cobra Effect.
- EssayAI Agents Inherit Your Incentives
How this element propagates into automated systems.
- EssayIncentives Under Crisis
How this element behaves under pressure.
- CourseIncentives 101
The free ten-part primer on reading a payoff structure.
- ReferenceThe Periodic Table of Human Behavior
The full 1,267-element map this page belongs to.
Questions about Counterfactual Testing
- What is Counterfactual Testing?
- Counterfactual Testing is testing model behavior on hypothetical alternate inputs. It sits in the Incentives dimension (INC) of the Human Behavior Taxonomy™ as element HBT-INC-0081, within the Risk family. The core principle: testing model behavior on hypothetical alternate inputs. In incentive terms, it matters because it changes the payoff people perceive before they choose — which means it can be designed for, or exploited.
- What is an example of Counterfactual Testing?
- Bias detection. The Incentives Lab catalogs everyday, organizational, and historical instances of this element on its Human Behavior Taxonomy™ page (HBT-INC-0081).
- How is Counterfactual Testing exploited?
- Bias detection.
- How do you design around Counterfactual Testing?
- Run counterfactuals on consequential models continuously.
- Which behavioral dimension does Counterfactual Testing belong to?
- Counterfactual Testing is classified in the Incentives dimension (INC) of the Human Behavior Taxonomy™, family "Risk", class "AI-Behavioral Coupling". Its permanent identifier is HBT-INC-0081 and its evidence grade is C.