AI Accountability
Pre-action constraints, liability architecture, and safety systems for AI in high-stakes operations.
Research focuses on how to design AI systems that are accountable before they act—not just auditable after harm occurs—drawing on industrial safety principles, constitutional design patterns, and operational risk management frameworks.
Core Accountability Frameworks
Pre-Deployment Rule Sovereignty
Default to Hold
Stop Work Authority
Architecture of Refusal
AI Accountability Frameworks
The AI Governance Gap
The Need for a Safety Brake
Mandatory Human Re-entry
The Watchdog Paradox
Velocity Over Capacity
AI Says 'Optimal' - Reality Fails
Related Tools
Constitutional Engine
Design pattern for embedding governance rules in AI systems
View in Tools →Architecture of Refusal
Framework for designing systems that refuse harmful operations
View in Tools →Industrial Safety Architecture
Applying industrial safety principles to AI systems design
View in Tools →Calvin Convention - Contractual Framework
Contractual structure for AI accountability in operations
View in Tools →