govrn.ai

01 / 15

TITLE

The standard for enterprise AI governance, powered by a deterministic measurement engine.

See it. Prove it. Run it.
One record, two lenses: govrn (the standard) and Aperture (the engine).
The standard sees it and proves it; the Aperture engine runs it.
Confidential investor material - June 2026.

Figures throughout are illustrative placeholders; product status described honestly - built vs. next.

02 / 15

PROBLEM

Every enterprise is spending real money on AI it cannot see

AI adoption outran AI governance, and the bill is already arriving in the dark.

Boards and regulators ask one question - is your AI governed, and can you prove it? Almost no one can answer.
Shadow AI runs unattributed across the estate, with no dollar figure attached.
The AI a company actually ships - its product AI - is the part nobody is reviewing.
Self-certification is not evidence; an auditor, board, or regulator won't accept it.

03 / 15

WHY NOW

The forcing function is regulation, not a vendor roadmap

The clock is already running, and it is set by law rather than by anyone's release cycle.

EU AI Act is in force - penalties to 7% of global turnover.
AI regulation quadruples to ~75% of world economies by 2030 (Gartner).
Board AI-risk oversight jumped 16% to 48% of the Fortune 100 in one year (EY).
Regulatory compliance is the #1 barrier to GenAI - 38% (Deloitte).
BCG draws 40%+ of 2025 revenue from AI/tech and is itself ISO/IEC 42001 certified - validation from the top.

04 / 15

See it. Prove it. Run it.

govrn.ai is the category answer - one record read through two lenses for two different buyers.

See + Prove (govrn, the standard): for the CISO and the board.
Run (Aperture, the engine): for the CTO and the CFO.
One record, two lenses, one source of truth - no reconciliation, no second system.
The interface carries evidence, not a verdict - posture is rendered from the same attributed record the engine maintains.

Cross-link: 02-the-standard-and-engine.html.

05 / 15

PRODUCT - STANDARD

The standard: govrn is live

A productized assessment a board, auditor, and regulator can accept - never self-certified.

Three-lens assessment: technology, security, and AI.
Six-framework crosswalks: NIST AI RMF, ISO/IEC 42001, EU AI Act, OWASP, ATLAS, CSF 2.0.
Independent attestation - the body that builds is never the body that attests.
Product-AI review: the AI a company ships, which is what a regulator cares about most.

06 / 15

PRODUCT - ENGINE

The engine: Aperture is built and tested

A deterministic measurement engine with no model in the path - reproducible and audit-defensible.

No model in the measurement path - every number is reproducible.
Pointers not payloads - metadata-only, enforced; the engine never holds the data.
Shadow AI with a dollar figure; attributed cost and usage across the estate.
Honest status: connectors are built but have NOT run against a live API. The $121k / 22%-shadow demo is the real engine on synthetic seed data (Meridian Software is a PRNG seed) - modeled, never a customer result.
Not yet built: multi-tenant isolation, auth, RBAC, audit - that is the next build.

Say modeled until connected, measured only after. The seam (govrn-interface) is designed in spec, not yet code.

07 / 15

MOAT

Why we win - independence and the accumulating record

The durable asset is structural independence plus an attributed record that compounds with time.

The body that builds is never the body that attests - the attest arm stays separate from build/run, even inside one company.
Determinism is the credibility - no model in the measurement path means a verdict survives audit.
The accumulated, attributed record compounds; an attestation is valid only against a fresh record, so measurement must be continuous.
Evidence stays portable - sovereignty, not lock-in. Stickiness is trust, not a trap.

08 / 15

MARKET

A regulated market expanding on a fixed clock

Governance spend is small today and structurally forced to grow under regulation.

AI-governance platform spend: $492M to $1B+ by 2030 (Gartner).
75% of world economies under AI regulation by 2030, quadrupling (Gartner).
Board AI-risk oversight: 16% to 48% of the Fortune 100 in one year (EY).
Demand is non-discretionary - the buyer cannot opt out of the clock.

Cross-link: 03-market.html.

09 / 15

BUSINESS MODEL

License the standard, keep the engine running

Deliver coupled, renew decoupled - the recurring dollar lands on the attested run-state.

Three things licensed: the process, the product, and the brand rights to govrn.ai.
Cost meter is the hook - fastest yes; managed run is the body; attestation is the true annuity a CISO can't casually churn.
An attestation is valid only against a fresh record - continuous measurement is required, not optional.
Services in the front, software underneath - assessment lands the account, the running engine is the relationship.
All pricing figures illustrative, pending Richard + counsel.

Cross-link: 04-business-model.html.

10 / 15

GO-TO-MARKET

Kelly is the first channel; Microsoft is the horizon

De-risked distribution now, with a senior-Microsoft path the team already holds.

MCG Kelly Services = the first channel - de-risked distribution into real enterprise accounts. Not the pitch target, not the destination.
Entity-first: stand up the company - entity, team, raise - before any partner conversation.
Microsoft is the horizon; the team holds senior-Microsoft-executive relationships through CTO Chris Bennett.
The exit profile: a defensible standard + an enterprise channel + a clock the buyer can't stop = a licensing or acquisition outcome.

11 / 15

TEAM

The bench is the de-risking

The people normally hired after funding are already at the table.

Richard Taubin - Founder: the standard, the brand, GTM.
Joe Saba - CAIO: the Aperture engine.
Chris Bennett - CTO: enterprise infrastructure, Azure, top-tier senior-Microsoft-executive relationships.
Roberto Lleras - CSO: data science and AI, AI-genomics/data reach.
Sean Barlow - full-stack and AI engineering.
Complete the wheel: an SME repository of the team's combined enterprise expertise is incoming; the next update names the full engine team.

Cross-link: 05-team.html.

12 / 15

TRACTION

What is real, stated honestly

We over-claim nothing - here is exactly what exists today.

Built and tested: the Aperture engine - deterministic, metadata-only, passing test suite, no model in the measurement path.
Live: the govrn standard - three-lens assessment, six-framework crosswalks, independent-attestation doctrine.
Demonstrated: the real engine on synthetic seed data ($121k / 22% shadow) - modeled, not a customer result.
Pre-MVP on live data: connectors built, never run against a live API; no multi-tenant isolation, auth, RBAC, or audit yet.
Designed, not yet code: the govrn-interface seam (markdown spec).

13 / 15

ROADMAP

Entity-first, then connect, then license

Stand up the company, prove it on one live account, then license at enterprise scale.

Now: stand up the entity, retain the founding bench, register the company.
Next build: connect a live API, harden to enterprise grade - multi-tenant isolation, auth, RBAC, audit; ship the govrn-interface seam as code.
Then: land the first design-partner accounts and move from modeled to measured.
Then license the process, product, and brand - delivered through Kelly as the first channel.

Cross-link: 06-roadmap.html.

14 / 15

THE ASK

The raise

Capital to turn a proven architecture into a licensable, category-defining standard.

$[ amount ] (illustrative - to be set).
Stand up the entity and retain the founding bench.
Harden the engine to enterprise grade and connect the first live account.
Land the first design-partner accounts - the bridge from modeled to measured.
All figures illustrative, to be finalized by Richard + counsel.