Enterprise AI Doctrine
AI Notes, Commentary and Synthesis
This site is a working collection of thoughts, models and reflections on artificial intelligence, with a focus on strategy, operating models, architecture and real world implementation.
Content includes original doctrine pieces as well as curated analysis of publicly shared work by others. All third party material is clearly attributed to its original authors and sources. The purpose is structured synthesis and commentary, not republishing.
This is not a news feed. It is a place to slow down, separate signal from noise and document what matters in enterprise AI.
© Daniel Mash — All third party content, logos, images, trademarks and referenced materials remain the property of their respective owners.
This doctrine is organised around eight core operating models:
Core Doctrine Models
AI Decision Ownership Model
https://mash.tech/ai-is-not-a-model-problem-it-is-a-decision-design-problem/
Governance Friction Curve
https://mash.tech/governance-friction-curve-enterprise-ai/
AI Capital Velocity Model
https://mash.tech/ai-capital-velocity-model/
Production Readiness Ladder
https://mash.tech/production-readiness-ladder-enterprise-ai/
Organisational AI Debt Index
https://mash.tech/organisational-ai-debt-index/
AI Architecture Selection Matrix
https://mash.tech/ai-architecture-selection-matrix/
Human–Agent Leverage Model
https://mash.tech/human-agent-leverage-model/
AI Portfolio Heatmap
https://mash.tech/ai-portfolio-heatmap/
Enterprise AI Playbook
A structured board-level series exploring capital allocation, governance design and institutional operating models for AI transformation:
Chapter 1 – Adoption vs Production
https://mash.tech/enterprise-ai-playbook-chapter-1-adoption-vs-production/
Chatper 2 – Captial Allocation
https://mash.tech/enterprise-ai-playbook-chapter-2-capital-allocation/
Chapter 3 – Governance & Culture
Chapter 4 – Data Liquidity & Architecture Constraint
Chapter 5 – Measurement & Economic Accountability
https://mash.tech/enterprise-ai-playbook-chapter-5-measurement-discipline-economic-visibility/
Chapter 6 – AI Portfolio Strategy <working>
Chapter 7 – Scaling & Institutionalisation <working>
Chapter 8 – Leadership & Operating Model Design <working>
Chapter 9 – AI Risk & Regulatory Posture <working>
Chapter 10 – Organisational Redesign in the Age of Agents <working>
The AI Portfolio Heatmap: Governing AI Investment Like Capital, Not Experimentation
Most enterprises manage AI initiatives like experiments. Mature organisations manage them like capital portfolios. When AI projects multiply without structure, visibility declines. Budgets fragment. Governance
The Human–Agent Leverage Model: Redesigning Work Without Destabilising It
Enterprise AI conversations often polarise quickly. One side predicts workforce replacement.The other insists AI will remain assistive. Both framings are incomplete. The real question is
The AI Architecture Selection Matrix: Choosing the Right Level of Intelligence
Most enterprise AI failures begin with overengineering. Teams default to the most sophisticated option available. Multi-agent systems, advanced orchestration layers and complex retrieval pipelines are
The Organisational AI Debt Index: The Hidden Constraint on Scale
Enterprises accumulate technical debt over time. They also accumulate organisational debt. AI exposes both. While technical debt can be refactored, organisational AI debt is more
The Production Readiness Ladder: Moving AI From Demo to Durable Infrastructure
Most AI initiatives look impressive in controlled environments. Few survive contact with production reality. The distance between demo and durable infrastructure is where enterprise AI
The AI Capital Velocity Model: Why Most Enterprise AI Spend Underperforms
Enterprise AI failure is rarely technical. It is economic. Boards approve AI budgets with strategic intent.Innovation teams launch pilots.Vendors demonstrate capability. Yet twelve months later,
The Governance Friction Curve: Why Enterprise AI Slows Down After the Pilot
Most AI initiatives do not fail in experimentation. They stall in institutionalisation. The pilot works.The demo impresses.The proof of concept shows promise. Then progress slows.
AI Is Not a Model Problem. It Is a Decision Design Problem
Most enterprise AI initiatives fail for the same reason. They focus on models. The real constraint is decision design. Boards are told they need LLMs,