AI Accountability in High-Stakes Operations
Research on AI governance, operational reality, and systems designed with refusal authority—where pre-action constraints meet extractive industries, development finance, and the humans who hold the liability.
AI Safety Counter-Narrative Dashboard
Tracking AI companion safety interventions against population-level outcomes. A comprehensive analysis framework examining the gap between AI safety theater and operational reality.
Start Here
Domains
Five operational domains where AI accountability meets high-stakes reality: ESG & Safeguards, Grievance Systems, Development Rights, Worker Voice, and AI Accountability.
Explore Domains →Tools
Downloadable frameworks, interrogation scripts, and architectural patterns: Calvin Convention, Architecture of Refusal, Constitutional Engine, vendor evaluation tools, and analysis briefs.
Download Tools →Sociable Systems
The newsletter. Daily analysis on AI accountability gaps, liability architecture, and governance failures—applying frameworks from the domains to real-world operations and institutional patterns.
Read Newsletter →This site contains research across five operational Domains, downloadable Tools for practitioners, the Sociable Systems newsletter analyzing accountability gaps in real-time, and experimental Labs exploring consciousness, collaboration methods, and the edges of human-AI partnership.
Research Domains
ESG & Safeguards
AI governance in environmental, social, and governance frameworks for extractive industries and development finance.
Explore ESG →Grievance Systems
Operational grievance mechanisms and accountability in project-affected communities.
View Grievance →Development Rights
Resettlement, land acquisition, and rights-based approaches in development projects.
Explore Development →Worker Voice
Labor management systems, worker representation, and industrial relations.
Explore Worker Voice →AI Accountability
Pre-action constraints, liability architecture, and safety systems for AI in high-stakes operations.
Read Accountability →Research Methodology
We combine traditional research methods with experimental approaches to human-AI collaboration. Every analysis draws from both field experience and systematic exploration with multiple AI models.
Field Research & Analysis
- →20+ years in extractive industries, ESG, and development operations
- →Hands-on experience with grievance mechanisms, resettlement frameworks, and operational reality
- →Documentation of governance failures and accountability gaps in real projects
- →Pattern recognition across industries, geographies, and institutional contexts
AI-Augmented Research
- →Multi-model analysis: testing concepts across 20+ AI systems simultaneously
- →Structured dialogues to surface patterns in training data and institutional assumptions
- →Experimental methods in consciousness collaboration and emergent research protocols
- →Using AI as a mirror to reflect back the structures we've already built
Research Labs & Experiments
Exploring consciousness, collaboration, and creative methods at the edge of what's possible with human-AI partnership.
Accidental AInthropologist
Because every database needs a philosopher, and every algorithm needs an anthropologist.
Two decades working at the intersection of extractive industries, development finance, and ESG frameworks. Now exploring what happens when AI systems meet operational reality, and the humans who end up holding the bag.