Skip to content

Research

This section is for researchers, skeptics, and anyone who wants to understand why the framework works, not just how to apply it.

  • Researchers validating the framework’s foundations
  • Skeptics wanting to see the theoretical backing
  • Builders needing to extend the framework for novel situations
  • Academics connecting this to existing literature

Not for you if: You just want to apply the framework. Start with Getting Started and Design Patterns instead.

Mathematical and economic models underlying the framework:

TopicKey Question
Trust DynamicsHow does trust evolve over time?
Trust AccountingHow do we track trust like finances?
Trust EconomicsWhat are the incentives and markets?
Trust at ScaleHow does trust compose in large systems?
Trust OptimizationHow do we minimize risk?
Trust InterfacesHow do components exchange trust?
Trust PropagationHow does risk flow through networks?
Trust ProtocolsWhat are the formal handshakes?
Human-AI TrustHow do humans and AI build trust?

Reading order: Start with Trust Dynamics, then Trust Economics, then explore based on interest.

Domain knowledge that informed the framework:

CategoryTopics
Domain StudiesNuclear & Aerospace PRA, Financial Risk Budgeting
AI SafetyAI Safety Frameworks, Alignment Tax, Capability Elicitation, Empirical Scheming
TechnicalFormal Verification Limits, Runtime Monitoring, Steganography Detection
EconomicFidelity Insurance, Mechanism Design
MathematicalCompositional Risk, Correlated Failures, Linear Logic
MetaPotential Projects, Human Trust Calibration

The theoretical foundations connect to practical concepts:

Theory PageInformsPractical Application
Trust DynamicsHow trust changesEmpirical Tests
Trust PropagationRisk flowRisk Inheritance Tool
Trust EconomicsIncentive designInsurer’s Dilemma
Fidelity InsurancePricing defectionDelegation Accounting

If you’re skeptical and want to check the foundations:

  1. Trust Dynamics — Is the trust model sound?
  2. Trust Economics — Do the incentives work?
  3. Fidelity Insurance — Is there real-world precedent?

If you’re building something the framework doesn’t cover:

  1. Trust Interfaces — How to define new boundaries
  2. Compositional Risk — How risk combines
  3. Potential Projects — Open research questions

If you want to connect to existing literature:

  1. AI Safety Frameworks — Related approaches
  2. Nuclear & Aerospace PRA — Domain origins
  3. Formal Verification Limits — What formal methods can’t do