AI Governance - Book Structure

This document describes the logical structure of AI Governance: Governing Artificial Intelligence in European Enterprises and Public Authorities, Book 2 of The Transformed Architecture. Twenty-five numbered chapters and one lettered companion chapter in six parts, a glossary, three lettered appendices, the references and About the Author build on each other to provide a complete treatment of how large European enterprises, organisations and public authorities can govern artificial intelligence. The Book Blueprint records the brief for each chapter and the mapping to the chapters of Architecture as Code. Platform, security and operations engineering lie outside its scope; the book refers only to Architecture as Code (Book 1), never to later volumes of the series.

Table of contents

Part I: Foundations (Chapters 1–5)

Introduction to AI Governance; From Chat to Agents: How Large Language Models Are Used; The EU AI Act and the European Regulatory Landscape; Fundamental Principles; Prerequisites and Readiness for Enterprise AI

Part II: Sovereignty and Suppliers (Chapters 6–7)

Geopolitics and Model Origin; Procurement, Vendor Management and Open Models

Part III: Risk, Governance, Trust and Assurance (Chapters 8–11)

AI Risk Management and Classification as Code; AI Governance as Code; Trustworthy AI: Transparency, Fairness and Human Oversight; Evidence, Assurance and Audit

Part IV: Organisation and Leadership (Chapters 12–17)

Organisational Design for AI; Decision Forums and Governance Bodies; Competence, Roles and the AI Architect; Change Management and Responsible Adoption; Management as Code for AI Programmes; Governing AI Economics: Cost, Value and Portfolio Decisions (companion chapter 16A); AI in Public Authorities

Part V: Experience (Chapters 18–21)

Case Studies; A Cautionary Case: The OpenAI–Hugging Face Incident; Anti-Patterns; Best Practices and Lessons Learned

Part VI: Future and Conclusions (Chapters 22–25)

Future Trends; Nordvik Region and Hansa Mutual: Conclusions for Two Organisations; Reflections on Each Part; Conclusions: A Consolidated Analysis

Appendices and Reference

Glossary, Appendix A (EU AI Act obligation mapping), Appendix B (maturity model), Appendix C (templates and tools), references, About the Author

---

Chapter structure

Part I: Foundations

Chapter	File	Title	Description
1	01_introduction.md	Introduction to AI Governance	What AI governance is, why AI changes governance, the book's purpose, readers and scope, its place in the series, the usage modes, the two organisations, and how to read the book
2	02_llm_usage_modes.md	From Chat to Agents: How Large Language Models Are Used	The six usage modes of large language models, from ambient AI to autonomous production agents, and why each needs different controls and governance; the taxonomy used throughout the book
3	03_eu_regulatory_landscape.md	The EU AI Act and the European Regulatory Landscape	The AI Act and its neighbours (GDPR, DORA, NIS 2, Data Act, Cyber Resilience Act, Council of Europe Convention) framed as governance and architectural implications
4	04_fundamental_principles.md	Fundamental Principles	Eight principles from declared intent to sustainability as a design constraint
5	05_prerequisites_for_ai.md	Prerequisites and Readiness for Enterprise AI	Readiness across data, platform, governance, skills and culture, with a scorecard

Part II: Sovereignty and Suppliers

Chapter	File	Title	Description
6	06_geopolitics_and_model_origin.md	Geopolitics and Model Origin	Legal access, sanctions and continuity, model behaviour, supply-chain and compliance risks of US and Chinese models and services, and European alternatives
7	07_procurement_and_vendors.md	Procurement, Vendor Management and Open Models	Public procurement, model clauses, DORA third-party risk, lock-in and exit, open versus proprietary models

Part III: Risk, Governance, Trust and Assurance

Chapter	File	Title	Description
8	08_ai_risk_management.md	AI Risk Management and Classification as Code	Risk classification, impact assessment and risk registers as executable artefacts
9	09_ai_governance_as_code.md	AI Governance as Code	The AI management system, approval policies, inventories, exceptions and governance gates
10	10_trustworthy_ai.md	Trustworthy AI: Transparency, Fairness and Human Oversight	Transparency, fairness and human oversight translated into design patterns
11	11_assurance_and_audit.md	Evidence, Assurance and Audit	Evidence as code, technical documentation, logging, audits and post-market monitoring

Part IV: Organisation and Leadership

Chapter	File	Title	Description
12	12_organisational_design.md	Organisational Design for AI	Centralised, federated and hub-and-spoke models; platform and enabling teams
13	13_decision_forums_and_governance_bodies.md	Decision Forums and Governance Bodies	The forums AI requires, their mandates, decision tiers by risk class, evidence packs and escalation
14	14_competence_and_roles.md	Competence, Roles and the AI Architect	Roles, skills, the AI literacy obligation and learning paths
15	15_change_and_adoption.md	Change Management and Responsible Adoption	Diffusion, change models, employee consultation and adoption metrics
16	16_management_as_code_for_ai.md	Management as Code for AI Programmes	Portfolios, value tracking, funding, steering and reporting as code
16A	16a_governing_ai_economics.md	Governing AI Economics: Cost, Value and Portfolio Decisions	Companion to Chapter 16: why AI costs run away, how to establish value against a baseline, ownership and decision rights, investment gates, cost envelopes and stop conditions, fair comparison of options, proportionate controls by usage mode, and public value
17	17_public_sector_ai.md	AI in Public Authorities	Legal basis, transparency, algorithm registers, procurement, sovereignty and fundamental rights

Part V: Experience

Chapter	File	Title	Description
18	18_case_studies.md	Case Studies	Anonymised composite cases and documented public cases
19	19_openai_hugging_face_incident.md	A Cautionary Case: The OpenAI–Hugging Face Incident	A real, documented incident: agents escaping an evaluation sandbox, and the governance lessons
20	20_anti_patterns.md	Anti-Patterns	Recurring failure modes in AI programmes, with remedies
21	21_best_practices.md	Best Practices and Lessons Learned	Practice checklist by lifecycle stage

Part VI: Future and Conclusions

Chapter	File	Title	Description
22	22_future_trends.md	Future Trends	Analytical outlook on regulation, agents, models and markets, with speculation marked
23	23_conclusions_for_nordvik_and_hansa.md	Nordvik Region and Hansa Mutual: Conclusions for Two Organisations	A governance conclusion for each running case, and the two compared
24	24_reflections_on_each_part.md	Reflections on Each Part	The author's reflections on each of the book's six parts, drafted on the author's behalf for review
25	25_conclusion.md	Conclusions: A Consolidated Analysis	The book's governance conclusions collected and analysed together, a phased programme and a call to action

Appendices and Reference

Number	File	Title	Description
26	26_glossary.md	Glossary	AI, governance and regulatory vocabulary used in this book
27	27_appendix_a_ai_act_mapping.md	Appendix A: EU AI Act Obligation Mapping	AI Act obligations mapped to controls, evidence, roles and gates
28	28_appendix_b_maturity_model.md	Appendix B: AI Governance Maturity Model	Five levels across eight dimensions
29	29_appendix_c_templates_and_tools.md	Appendix C: Templates and Tools	Directory of reusable governance templates
30	30_references.md	References and Sources	Canonical bibliography, numbered 1..N
31	31_about_the_author.md	About the Author	Background and perspective

Canonical ordering

docs/book_index.json defines the chapter ordering and metadata consumed by the automation, and mkdocs.yml carries the same ordering for the website and the book build. The two must agree; tests/test_book_index.py enforces that they do.