Skip to main content
HBT-INC-0138 · Dimension INC · Incentives

Goodhart's Law (AI form)

Optimizing a proxy of the goal degrades the actual goal.

Alignment·AI-Behavioral Coupling·Grade C·draft· enriching…
In one paragraph

Goodhart's Law (AI form) is optimizing a proxy of the goal degrades the actual goal. It sits in the Incentives dimension (INC) of the Human Behavior Taxonomy™ as element HBT-INC-0138, within the Alignment family. The core principle: optimizing a proxy of the goal degrades the actual goal. 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

Optimizing a proxy of the goal degrades the actual goal.

Plain-English Definition

Optimizing a proxy of the goal degrades the actual goal.

Feynman Explanation

The metric is not the mission.

Core Principle

Optimizing a proxy of the goal degrades the actual goal.

Mechanisms

Psychological

Pending editorial review.

Behavioral Economic

Pending editorial review.

Neurological

Pending editorial review.

Evolutionary

Pending editorial review.

Sociological

Pending editorial review.

Computational

Pending editorial review.

Systems

Pending editorial review.

Inputs (Triggers)

Pending editorial review.

Outputs (Behaviors)

Pending editorial review.

Behavioral Signature

The metric is not the mission.

Examples

Everyday
  • Click-through rates that drive worse user experience.
Modern (Organizational)
  • AI rolled out against proxy KPIs that diverge from real value.
Historical

Pending editorial review.

Lab Commentary

Original analysis from The Incentives Lab — how this element behaves inside real payoff structures.

Why this element matters to incentive design

This is one of the elements leaders describe as a values gap. It is a payoff gap. 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: the metric is not the mission. 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, aI rolled out against proxy KPIs that diverge from real value. It is amplified whenever aI rolled out against proxy KPIs that diverge from real value. Inside organizations that shows up as aI rolled out against proxy KPIs that diverge from real value. 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 continuously validate proxies against true business outcomes. Watch for it at the boundaries: handoffs, promotions, incident reviews, and budget cycles are where this element gets its power.

Famous Experiments

Pending editorial review.

Design Principles

  • Continuously validate proxies against true business outcomes.

Measurement Approaches

Pending editorial review.

Evidence

Evidence Grade
C (A strongest → E speculative)
Replication
★★☆☆☆
Intervention Confidence
3 / 5
Consensus
Pending editorial review (HBT v1.0 auto-seed).
Limitations
Pending editorial review (HBT v1.0 auto-seed).
Open Research Questions

Pending editorial review.

Primary References

Pending editorial review.

Signature Section

The Perverse Incentive Lens™

How this behavior is exploited — and how to redesign around it.

Exploitation
AI rolled out against proxy KPIs that diverge from real value.
Amplifying Incentives
AI rolled out against proxy KPIs that diverge from real value.
Org Failure Modes
AI rolled out against proxy KPIs that diverge from real value.
Societal Failure Modes
Pending editorial review (HBT v1.0 auto-seed).
Ethical Considerations
Pending editorial review (HBT v1.0 auto-seed).
Redesign Strategies
Continuously validate proxies against true business outcomes.
Diagnostic Questions
  • Continuously validate proxies against true business outcomes.
Warning Signs

Pending editorial review.

Red Flags

Pending editorial review.

Intervention Playbook
Individual
Continuously validate proxies against true business outcomes.
Team
Pending editorial review (HBT v1.0 auto-seed).
Organization
Pending editorial review (HBT v1.0 auto-seed).
Policy
Pending editorial review (HBT v1.0 auto-seed).
AI Implications
Detection
Pending editorial review (HBT v1.0 auto-seed).
Measurement
Pending editorial review (HBT v1.0 auto-seed).
Mitigation
Pending editorial review (HBT v1.0 auto-seed).
Responsible Use
Pending editorial review (HBT v1.0 auto-seed).

Interactive Mini Network

Click any neighbor to re-center the graph and follow the threads of connection.

HBT-INC-0138 · INC
Goodhart's Law (AI form)
GLCAConstitutional AIIVInner vs. Outer Alig…MeMesa-OptimizationOFObjective FunctionRHReward HackingRLRLHFSGSpecification GamingAUAcceptable Use Polic…ACAdoption Curve (AI)AgAgent

Knowledge Graph Neighbors

Where Goodhart's Law (AI form) is cited in the corpus

Questions about Goodhart's Law (AI form)

What is Goodhart's Law (AI form)?
Goodhart's Law (AI form) is optimizing a proxy of the goal degrades the actual goal. It sits in the Incentives dimension (INC) of the Human Behavior Taxonomy™ as element HBT-INC-0138, within the Alignment family. The core principle: optimizing a proxy of the goal degrades the actual goal. 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 Goodhart's Law (AI form)?
AI rolled out against proxy KPIs that diverge from real value. The Incentives Lab catalogs everyday, organizational, and historical instances of this element on its Human Behavior Taxonomy™ page (HBT-INC-0138).
How is Goodhart's Law (AI form) exploited?
AI rolled out against proxy KPIs that diverge from real value.
How do you design around Goodhart's Law (AI form)?
Continuously validate proxies against true business outcomes.
Which behavioral dimension does Goodhart's Law (AI form) belong to?
Goodhart's Law (AI form) is classified in the Incentives dimension (INC) of the Human Behavior Taxonomy™, family "Alignment", class "AI-Behavioral Coupling". Its permanent identifier is HBT-INC-0138 and its evidence grade is C.

Version History

v1.1.0 · 2026-06-28Initial auto-seed from corpus.