Keyboard shortcuts

Press or to navigate between chapters

Press S or / to search in the book

Press ? to show this help

Press Esc to hide this help

The Adversarial Platform Canvas

A reusable decision record for moving from ecosystem harm to proportionate intervention.

Status: working frameworkUse: every chapter and case

How to use it

Complete the canvas before selecting a detector. Cite evidence, mark assumptions, name an owner, and preserve alternatives. Revisit it after incidents, policy changes, major model changes, or attacker adaptation.

#DimensionRequired output
1EcosystemProtected exchange, community, resource, or experience and its boundaries
2ActorsDirect users, intermediaries, affected non-users, operators, and adversaries
3IncentivesBenefits, costs, constraints, substitutes, and externalities for each actor
4HarmObservable undesirable outcomes, affected parties, magnitude, and reversibility
5PolicyAllowed, limited, discouraged, restricted, and prohibited behavior
6ObservabilityNecessary, lawful, reliable, and retained evidence with lineage
7DetectionInference methods and the claims each can actually support
8ConfidenceCalibration, base rate, manipulability, ambiguity, and disagreement
9UrgencyHarm accumulation and the detection/decision latency budget
10InterventionTarget, scope, severity, duration, reversibility, and expected effect
11Error costsFalse-positive, false-negative, friction, and collateral costs by population
12RecourseNotice, explanation, review, appeal, correction, and audit trail
13EconomicsAttacker and defender cost changes, scale limits, and displacement
14AdaptationExpected probing, evasion, imitation, migration, and next measurement

Decision record

Decision:
Ecosystem and harm:
Policy authority:
Evidence and lineage:
Inference and confidence:
Latency budget:
Chosen intervention:
Why proportionate:
Error and subgroup risks:
Recourse and rollback:
Success and guardrail metrics:
Expected adaptation:
Owner and review date:

Quality checks

  • Would the action still be justified if the actor were human rather than automated?
  • Does the evidence support the intervention target, or only a related request/account/device?
  • Is a less invasive or more reversible action sufficient?
  • Can a legitimate user understand the rule and recover from error?
  • What observation would prove the intervention ineffective or harmful?