AI Registry: Register Every AI System With Two Named Owners

An AI registry lists every AI system your company runs in production or pilot, with two named owners on each. A red AI Register Completeness score proposes the AI Register Build, which the AI governance lead runs for 3 weeks. The run closes when the outcome is met and an independent re-score lands at amber or better.

What an AI Registry Means in the AI Deployment Canvas

An owner who has just finished the free AI Readiness Assessment on vwcg.app may want to read how a full operating system handles AI governance. That tool scores six dimensions: AI Strategy, Data Readiness, Infrastructure, Talent and Skills, Governance, and Culture. It is a separate instrument, and its result does not feed any score in the module diagnostic below.

The VWCG OS handles AI governance inside its AI Deployment Canvas module. The diagnostic for that module scores five dimensions on a 0 to 5 scale against written anchors. Each dimension carries an equal weight of 0.20 in the composite score. One of those five dimensions, AI Register Completeness, scores the register itself.

The other four dimensions are Readiness Gate Discipline, Scoring Grid Coverage, Pipeline Stage Control, and Audit Cadence and Response. Each one routes to a remediation playbook when it reads red. The sections below stay with the register, because a red register score points to the AI Register Build.

In this module the registry carries the name AI Register. It records every AI deployment that runs inside six other modules of the system. Those modules are SOP Codex, Client Success Loop, Sales Velocity Engine, Agile Capital Allocation, People and Culture Analytics, and Cyber, Data Privacy, and Security.

What the AI Register Completeness Dimension Measures

The dimension asks whether every AI deployment sits in the register with a full set of fields. Each entry needs a unique id, the model and provider, the data sources, and a risk score. It also needs a Model Owner, a Business Owner, a status, and an incident log. The dimension also checks that board visibility of the register stays current.

The written anchors describe three points on the scale. A score of 1 means different teams deploy AI tools with no central record. At a score of 3, a register exists, but its entries are stale or its ownership is single-sided. Score 5 requires a complete and current register, with dual ownership on every entry.

The top anchor adds two more conditions to that picture. Someone maintains the incident logs, and board review happens quarterly. In balanced scorecard terms, the anchors turn a broad governance goal into an observable measure that an owner can check.

How a Red Score Becomes a Route

The module reads scores against three bands on the 0 to 5 scale. Red covers the low end, amber covers the middle, and green covers the top. The bands cover the full scale with no gaps or overlaps.

A red AI Register Completeness score routes to the AI Register Build playbook. An amber score routes to the AI Governance Monitoring playbook, which runs a 30-day pulse check for each amber dimension. A green score holds, and the Core Execution module monitors it quarterly.

A route that fires from a scored dimension stays a proposal until leadership confirms it. Leadership confirms it at the next monthly governance review or the Core Execution Phase 3 review, whichever comes first. The 3-week duration of the AI Register Build starts at that confirmation, not at the scoring.

A color changes only when a score clears the band edge by 0.25 points or holds for two consecutive scorings. In PDCA terms, this rule decides when a check result may change the next act. The system scores module instruments quarterly, in weeks 11 to 12 of the Core Execution cycle.

What Opens the AI Register Build

Three conditions open the playbook, and any one of them is enough. AI Register Completeness scores red, someone discovers AI tools outside governance, or the board requests AI visibility. A trigger that arrives while the playbook is already open goes into the log of the open run.

Two other modules can also send work into the register. A red Churn Predictor Governance score in the Client Success Loop sends the churn predictor and the QBR generator to the register with both owners. A red Lead Scoring and AI Governance score in the Sales Velocity Engine does the same for lead scoring and the proposal generator. Those two systems also receive quarterly bias audits.

What the AI Register Build Does

The playbook inventories every AI system in production or pilot across all modules. The scope runs from Draft-With-AI in the SOP Codex, the module for standard operating procedures, to threat detection in Cyber, Data Privacy, and Security. It also covers team-level tools.

Every entry carries the full field set with dual ownership. The team marks uncertain incident histories rather than inventing them. In Lean terms, the inventory works like a current state map, a record of the AI systems that actually run today.

The AI governance lead owns the playbook with a data or machine learning counterpart. Its stated duration is 3 weeks from confirmation. The outcome has three parts that must all hold at the end of the run.

First, every known AI deployment sits in the register with dual ownership and status. Second, the board has seen the register. Third, a quarterly review calendar exists.

Dual Ownership on Every Entry

The Model Owner holds technical accountability for the deployment. A Business Owner holds impact accountability for the same deployment. The register names both people on every entry.

In RACI terms, the register splits accountability into two named scopes, technical and impact, rather than leaving it with one person. The score 3 anchor names single-sided ownership as a mark of a register that falls short. The score 5 anchor requires dual ownership on every entry, so two named owners is a scored condition.

How the Run Closes or Escalates

Every playbook follows one lifecycle: open, running, met or missed, escalated, closed. An outcome met, with an independent re-score at amber or better, closes the run. Someone other than the owner re-scores the dimension.

An outcome missed at the end of the 3 weeks escalates to the owner’s manager. The miss also goes onto the next Core Execution quarter planning agenda. Each owner runs at most three red playbooks at once, and further reds queue.

The queue ranks those reds by how many modules the red dimension routes into. When a red playbook closes, the owner’s next queued red opens. In theory of constraints terms, the cap and the queue make owner capacity an explicit limit on how many red playbooks run at once.

What Keeps the Register Current

When the AI Register Build outcome is met, any standing duty it set up continues under the AI Governance Monitoring playbook. That playbook reviews the AI Register monthly for stale statuses, ownership gaps, and unlogged incidents. It also confirms audit cadence for every deployment.

The same standing playbook produces quarterly reports on incident histograms and override trends. It adjusts pipeline mode when a reading crosses an AERI tier boundary in the Intelligent Foundations module. New AI proposals must arrive with grid scores before any funding conversation in Agile Capital Allocation.

The AI governance lead owns that standing playbook, and it does not count toward the cap of three red playbooks. Its outcome reads simply: register current, mode matches AERI, roll-ups produced, no overdue audit, and Core Execution fed. Owners who want to read about outside help with this kind of work can start with a piece on organizational consulting.

For a broader read on readiness, see the vwcg.app article on an AI readiness assessment. The free tool on vwcg.app stays separate from the AI Deployment Canvas diagnostic.

Frequently Asked Questions

What is an AI registry in the AI Deployment Canvas?

In the AI Deployment Canvas, the AI registry is the AI Register. It records every AI deployment across the SOP Codex, Client Success Loop, Sales Velocity Engine, Agile Capital Allocation, People and Culture Analytics, and Cyber, Data Privacy, and Security modules. Each entry names a Model Owner and a Business Owner, and board visibility stays current.

What does the AI Register Completeness dimension measure?

AI Register Completeness checks that the register holds every AI deployment with a unique id, model and provider, data sources, a risk score, a Model Owner, a Business Owner, a status, and an incident log. It also checks current board visibility. It is one of five dimensions, each scored 0 to 5 with a weight of 0.20.

When does the AI Register Build open?

The AI Register Build opens on any of three triggers: a red AI Register Completeness score, AI tools discovered outside governance, or a board request for AI visibility. A route from a scored dimension stays a proposal until leadership confirms it. The stated 3-week duration of the playbook starts at that confirmation.

What does each AI registry entry carry?

Each entry carries a unique id, the model and provider, the data sources, a risk score, a Model Owner, a Business Owner, a status, and an incident log. The Model Owner holds technical accountability, and the Business Owner holds impact accountability. Uncertain incident histories are marked rather than invented during the build.

Who owns the AI Register Build and what is the outcome?

The AI governance lead owns the playbook with a data or machine learning counterpart, over 3 weeks. The outcome requires every known AI deployment registered with dual ownership and status, a board that has seen the register, and a quarterly review calendar. An outcome met, with an independent re-score at amber or better, closes the run.

How does World Consulting Group help an owner take the next step?

World Consulting Group implements the VWCG OS, including the AI Deployment Canvas diagnostic. The first step of an engagement is a scoring pass of the five AI Deployment Canvas dimensions against the written anchors. That pass shows whether AI Register Completeness reads red, amber, or green, and which playbook the system proposes for leadership to confirm.

The AI Readiness Assessment is one of the free assessment tools on vwcg.app, built by World Consulting Group. Start the free assessment

Kamyar Shah has led 650+ consulting engagements, including fractional COO, fractional CMO, executive coaching, and strategic advisory, producing over $300M in client impact across companies in the $1M-$50M range. He built the VWCG Strategic Assessment from the same diagnostic frameworks he uses in paid engagements.