The library
Whitepapers on runtime AI governance
Five papers on the control problem enterprise AI actually has: agents that can act, not just answer. Every one is published in full, with no form and no gate.
From Monitor-First to Default-Deny
How to prevent unapproved tool execution in production
Read the paper →Sensitive Knowledge and Retrieval Governance
How to stop AI agents from seeing more than they should
Read the paper →Discover Before You Govern
Why AI governance starts with inventory, not policy
Read the paper →Verifiable Decision Provenance
Why AI governance needs proof rather than logs
Read the paper →Policy-Bounded Business Domains for AI Agents
How to keep AI agents inside approved business lanes
Read the paper →The instrument
The evaluation scorecard
Each paper closes with the questions to put to a vendor. Assembled, they are a 25-question scorecard for evaluating any AI governance platform, including ours.