Engagement Models

PreMetric engagements are structured, bounded, and decision-specific. Each audit or reassessment is initiated at a defined decision point and concludes with a documented AI Decision Record. PreMetric does not provide continuous monitoring, ongoing oversight, or retainer-based access.

Five engagement types

01AI Initiative Audit
02AI Vendor Enterprise Readiness Audit
03Investor AI Diligence Audit
04Portfolio AI Exposure Audit
05Decision Assurance Review

Each audit or reassessment produces a structured AI Decision Record. Point-in-time only — not retainer-based.

Defined outputs across every engagement

Every PreMetric engagement produces a defined set of outputs grounded in assessment of the AI system, technical evidence, vendor claims, and institutional context. These are not formal certifications, standalone benchmark reports, or advisory opinions — they are structured institutional records designed to support accountable decision-making by boards, executives, investors, procurement teams, and governance bodies.

01

Decision record

02

Evidence chain

03

Deployment readiness assessment

04

Governance and accountability analysis

05

Capital exposure assessment

06

Proceed / modify / pause / stop recommendation

ProceedModifyPauseStop
07

Decision triggers for reassessment

Five engagement types

PreMetric offers four pre-deployment AI audits and one reassessment engagement, each calibrated to a specific decision context. Each engagement is bounded, documented, and decision-specific. All follow the same documented methodology and produce the same defined output set.

01
AI Initiative Audit

AI Initiative Audit

Enterprises · Boards · Executive teams

A focused audit of a single AI initiative, AI product, vendor selection, deployment proposal, or AI-enabled business case before approval, procurement, capital commitment, or deployment. PreMetric examines the system, evidence, assumptions, controls, deployment context, institutional exposure, and decision rationale to determine whether the initiative should proceed, be modified, be paused, or be stopped.

Scope

  • One AI initiative, product, vendor selection, deployment proposal, or AI-enabled business case
  • AI system purpose, vendor claims, model assumptions, data assumptions, and performance evidence
  • Control environment, human oversight, escalation pathways, and deployment dependencies
  • Capital exposure, accountability structure, and risk boundaries
  • Governance readiness and regulatory exposure

Deliverables

  • AI Decision Record with documented rationale
  • Evidence chain supporting the recommendation
  • Assumption, control, and exposure summary
  • Proceed / modify / pause / stop recommendation
    ProceedModifyPauseStop
  • Decision triggers for reassessment

Pricing reflects initiative complexity and capital exposure.

02
AI Vendor Enterprise Readiness Audit

AI Vendor Enterprise Readiness Audit

AI companies · Vendors · Product teams

For AI companies preparing for enterprise buyers, procurement scrutiny, regulated-sector customers, strategic partnerships, or investor diligence. This audit strengthens the evidence chain around product value, deployment assumptions, governance, accountability, risk boundaries, and buyer approval logic.

Scope

  • Product value claims and deployment assumptions
  • Governance and accountability structure
  • Risk boundaries and regulatory posture
  • Enterprise procurement and buyer approval readiness

Deliverables

  • Evidence chain supporting enterprise and investor review
  • Deployment readiness assessment
  • Governance and accountability analysis
  • Identified gaps and prioritised remediation
  • Decision record for enterprise or investor use

Pricing reflects product complexity and target deployment context.

Enterprise Readiness for AI Companies
03
Investor AI Diligence Audit

Investor AI Diligence Audit

Funds · Family offices · Acquirers · Transaction teams

For funds, family offices, acquirers, and transaction teams evaluating AI-related value claims before committing capital. This audit assesses whether the AI upside is credible, deployable, and defensible.

Scope

  • AI value claims embedded in the investment thesis
  • Deployment assumptions and credibility assessment
  • Capital exposure and downside risk analysis
  • Governance, regulatory, and accountability exposure

Deliverables

  • Investment decision record
  • Evidence chain supporting or qualifying AI value claims
  • Capital exposure assessment
  • Proceed / modify / pause / stop recommendation
    ProceedModifyPauseStop
  • IC-ready documentation

Pricing reflects transaction size, AI complexity, and timeline.

AI diligence for investors
04
Portfolio AI Exposure Audit

Portfolio AI Exposure Audit

PE firms · Venture funds · Corporate portfolios

A structured snapshot of selected portfolio companies or AI-dependent initiatives, identifying where AI creates value, where assumptions are weak, and where governance, procurement, deployment, or regulatory risk may exist.

Scope

  • Selected portfolio companies or AI-dependent initiatives
  • Value creation and assumption validity per company
  • Cross-portfolio governance and risk pattern analysis
  • Procurement, deployment, and regulatory exposure mapping

Deliverables

  • Per-company decision records
  • Portfolio-level capital exposure assessment
  • Evidence chain for identified value and risk positions
  • Governance and accountability analysis across holdings
  • Prioritised reassessment triggers by company

Pricing based on portfolio scope and number of companies audited.

05
Decision Assurance Review

Decision Assurance Review

Any organisation with a previously reviewed AI initiative

An episodic reassessment when an AI initiative materially changes, enters a new market, expands in scope, moves toward deployment, or faces new governance or regulatory scrutiny. Each review revalidates or revises the original decision record.

Scope

  • Material changes in initiative scope, capability, or context
  • New market entry or regulatory environment
  • Movement from development toward deployment
  • New governance, board, or regulatory scrutiny

Deliverables

  • Updated decision record confirming or revising prior determination
  • Revised evidence chain reflecting material changes
  • Updated deployment readiness assessment
  • Revised proceed / modify / pause / stop recommendation
    ProceedModifyPauseStop
  • New decision triggers for future reassessment

Triggered by defined decision events, not by calendar.

How PreMetric audits are bounded

PreMetric audits are structured, bounded, and point-in-time. They examine the system, evidence, assumptions, controls, deployment context, and institutional exposure behind an AI initiative. The output is an AI Decision Record, not a certification, implementation plan, or ongoing monitoring function.

Audit output

PreMetric produces AI Decision Records, evidence chains, findings, assumption and exposure summaries, and proceed / modify / pause / stop determinations. Formal certifications, conformity assessments, and legal compliance opinions remain separate downstream processes.

Deployment role

PreMetric does not build, deploy, operate, or monitor AI systems. The audit is conducted before commitment, where system evidence, controls, exposure, and institutional readiness can still be assessed objectively.

Engagement model

PreMetric audits are bounded and decision-specific. Post-deployment reassessments are reserved for material changes in scope, risk, use, evidence, or deployment context.

Technical evidence

Benchmark results, validation materials, vendor assurance documentation, and certification-readiness indicators may be assessed as audit inputs. They are not the output of the engagement unless incorporated into the AI Decision Record.