D
Loading
Available
The Three-Body Problem of Classified AI
Governance Architecture for AI in Federal Authorization Boundaries
Classification, compliance, and developer velocity form a chaotic three-body system. No two of those forces stay stable without the third, and no manual process holds the orbit. Volume II is the federal answer.
Buy
Purchase links will appear here at launch.
Inside this volume
The Mosaic Effect
Permissible fragments that an AI can assemble into impermissible disclosures by combining context across compartments. Each piece is cleared; the combination is not. The Mosaic Effect is why classification reviewers exist, and why your agent needs one too.
Table of contents
What you'll read inside Volume II
- 01The Framework in Brief
- 02Prologue: Two Rooms, One Problem
- 03Chapter 1: The Three-Body Diagnosis
- 04Chapter 2: The Formal Adversary
- 05Chapter 3: Non-Predictability From Initial Conditions
- 06Chapter 4: No Closed-Form Solution
- 07Chapter 5: The Authorization Threshold
- 08Chapter 6: Residual Risk by Design
- 09Chapter 7: The Power Persuaders and the Polish Trap
- 10Chapter 8: Codebase Preconditions
- 11Chapter 9: Knowledge Base Integrity
- 12Chapter 10: Model Supply Chain
- 13Chapter 11: Governance of Noise
- 14Chapter 12: Drift and Decay
- 15Chapter 13: Design Principles for Operational Trust
- 16Chapter 14: Identity, Intent, and Chain of Custody
- 17Chapter 15: The Delegation Token and Activity Record
- 18Chapter 16: The Agentic RAG Architecture
- 19Chapter 17: The Governance Engine
- 20Chapter 18: Specialized Agents and Routing
- 21Chapter 19: Multi-Developer Agent Orchestration
- 22Chapter 20: Federated Trust Across Boundaries
- 23Chapter 21: Edge-Native Autonomy
- 24Chapter 22: Harness Theory
- 25Chapter 23: P1-P8 as Harness Constraints
- 26Chapter 24: The Three Failure Modes
- 27Chapter 25: From Governance to Evidence
- 28Chapter 26: The Evidence Stream
- 29Chapter 26b: The Last Honest Witness
- 30Chapter 27: Policy-as-Code and Risk Registers
- 31Chapter 28: The Authorization Decision Framework
- 32Chapter 29: The AO's Operating Picture
- 33Chapter 30: Trust But Verify
- 34Chapter 31: Policy-Gated Autonomy for AI Agents
- 35Chapter 32: Automated Compliance Evidence
- 36Chapter 33: Automated ATO Pre-Submission Compliance Scanning
- 37Chapter 34: Memory Injection Governance
- 38Chapter 35: The Dream Phase
- 39Chapter 36: The Loophole Test
- 40Chapter 37: Governance Gates
- 41Chapter 38: Trust-But-Verify for Cross-System Agents
- 42Chapter 39: The Mosaic Effect
- 43Appendix A: Properties Reference
- 44Appendix B: Adversary Classes A1-A7
- 45Appendix C: Standards Alignment
- 46Appendix D: Configuration Reference Compendium
- 47Appendix E: Companion Volume Navigation
- 48Appendix F: Index
- 49Appendix G: Blast Radius and Trajectory, Two Operational Instruments
- 50Appendix H: Implementation Economics
- 51Appendix I: The Vendor-Neutral Harness
- 52Appendix J: Constitutional and Legal Considerations
Back cover
