Skip to main content
The Incentives Lab
HBT-INC-0113 · Dimension INC · Incentives

Eval Drift

Evaluation suites that no longer reflect real-world conditions.

Workflow·AI-Behavioral Coupling·Grade C·draft· enriching…

Scientific Definition

Evaluation suites that no longer reflect real-world conditions.

Plain-English Definition

Evaluation suites that no longer reflect real-world conditions.

Feynman Explanation

Your benchmarks aged out of relevance.

Core Principle

Evaluation suites that no longer reflect real-world conditions.

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

Your benchmarks aged out of relevance.

Examples

Everyday
  • Benchmark scores rising while real usage quality declines.
Modern (Organizational)
  • Quality erosion masked by stale measurement.
Historical

Pending editorial review.

Famous Experiments

Pending editorial review.

Design Principles

  • Periodically refresh evals against current production data.

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
Quality erosion masked by stale measurement.
Amplifying Incentives
Quality erosion masked by stale measurement.
Org Failure Modes
Quality erosion masked by stale measurement.
Societal Failure Modes
Pending editorial review (HBT v1.0 auto-seed).
Ethical Considerations
Pending editorial review (HBT v1.0 auto-seed).
Redesign Strategies
Periodically refresh evals against current production data.
Diagnostic Questions
  • Periodically refresh evals against current production data.
Warning Signs

Pending editorial review.

Red Flags

Pending editorial review.

Intervention Playbook
Individual
Periodically refresh evals against current production data.
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-0113 · INC
Eval Drift
EDAgAgentAWAI-First Workflow De…CDConcept DriftCWContext WindowDDData DriftEmEmbeddingEFEvaluation Framework…ILIn-Context LearningLBLatency BudgetMDModel Distillation

Knowledge Graph Neighbors

Version History

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