Playbook · White paper · Newsroom
Insights
Operator-written. Posted when defensible.
What we publish.
-

01 · Playbook
The AI Operating Playbook
Could your AI estate survive a board review tomorrow?
12 questions, 10 seconds each. Then the playbook: 32 pages, 4 operating instruments, free.
Open the AI Operating Playbook -

02 · White paper, in French
AI governance is an architecture problem.
20 pages on why generative AI projects fail, and what the few that succeed build first.
Open the white paper -
03 · Newsroom
Field briefings for AI operators.
Operator notes on AI governance, architecture and regulation: running the AI operating layer with regulated‑grade rigor, from strategy to execution.
Open the Newsroom
Reference guides
One text, one buyer question, dated primary sources.
- NIST AI RMF implementation, control by control 15 governance controls mapped to the NIST subcategories, with their evidence and the US status.
- Enterprise AI agents implementation, phase by phase 4 phases, the controls on every agent and the OWASP agentic risks.
- AI in production for the middle market Adoption data from NCMM, RSM, the Census Bureau and Eurostat, and the operating path.
- AI audit trail: definition and EU AI Act mapping The 9 fields of the record and Articles 12, 19 and 26.
- AI literacy and Article 4 of the EU AI Act The duty as reworded in July 2026, and the records that show it is met.
- FINMA Guidance 08/2024 on AI, expectation by expectation The 7 areas FINMA assessed, and the evidence for each.
- The EU AI Act for Swiss companies Scope, roles and dates for a company established in Switzerland.
- EU AI Act deployer obligations, control by control Articles 4, 26, 27 and 50, each with its governance control and its record.
- Automated individual decisions under the Swiss FADP Articles 21 and 22, the exceptions of paragraph 3 and the record that answers a review request.
- AI laws in the United States: NIST, Texas, Colorado No federal AI statute: the NIST framework, Texas and Colorado, dated in their primary sources.
- The Colorado AI Act from 1 January 2027 SB24-205, SB25B-004, SB26-189 and the court stay, with the records a deployer keeps.
- Texas HB 149, the Texas Responsible AI Governance Act In effect since 1 January 2026: the framework defense, the 60-day cure and the records it requires.
- Glossary of the AI operating layer Each term in one standalone sentence, with its primary sources, its date and a stable address.
AI in production, sector by sector.
The regulation attaching to AI differs by sector. The architecture that satisfies it does not change, only the policy it enforces.
Financial services
Model risk management and explainability obligations apply to every credit, underwriting or trading decision an AI system touches. See the financial services page
Healthcare
Clinical and administrative AI carries patient data and safety duties that make an audit trail a condition of deployment, not an add-on. See the healthcare page
Energy
Grid and infrastructure operators run AI against systems where a runtime failure has physical consequences, not only a compliance one. See the energy page
Luxury
Heritage brands run AI against client data and craft records that carry their own confidentiality standard, board-level and generational. See the luxury page
