EVF Diagnostic
EVF Diagnostic.
A structured governance diagnostic that shows what your organisation can prove before runtime enforcement begins.
EVF captures the execution boundary, evidence quality, risk signals, AI system inventory, and priority actions in a report your team can hand to auditors, counsel, leadership, or implementation partners.
EVF supports AI governance assessment, execution viability review, and audit-ready diagnostic preparation.
Not a questionnaire. An evidence-producing diagnostic.
What EVF does.
EVF identifies where governance breaks before OBEXGATE is deployed into the execution path.
It captures how AI systems are used, who owns the execution boundary, what evidence exists today, where legal and liability exposure may sit, and which gaps need action first.
The output is a diagnostic report with risk signals, priority actions, practitioner notes, and recommended next steps.
What EVF captures.
EVF captures organisation metadata, practitioner traceability, AI system inventory, system purpose, vendor/provider information, high-risk classification posture, and governance evidence across the organisation.
The canonical intake instrument supports up to 6 AI systems and 324 structured fields.
- Organisation and practitioner metadata
- AI system inventory and intended purpose
- Execution boundary ownership
- Incentive alignment
- Adoption pathway and mandate posture
- System complexity and interdependencies
- Evidence quality
- Adversarial robustness
- Drift and continuous validity
- Gate integrity under pressure
- Legal and liability exposure
- Navigation architecture for affected parties
EU AI Act reference domains.
For systems in scope, EVF includes per-system EU AI Act reference domains covering training data, technical documentation, logging, transparency, and technical robustness.
It also captures organisation-wide AI Literacy and Fundamental Rights Impact Assessment posture.
- Article 10: Training Data and Data Governance
- Article 11 and Annex IV: Technical Documentation
- Article 12: Logging Content Adequacy
- Article 13: Provider-to-Deployer Transparency
- Article 15: Technical Robustness and Cybersecurity
- Article 4: AI Literacy
- Article 27 reference posture: Fundamental Rights Impact Assessment
What the report gives you.
The EVF report turns intake findings into an actionable governance artefact.
It shows risk signal distribution, domain-level findings, priority actions, practitioner notes, and recommended next steps.
It is designed to support internal leadership review, audit preparation, implementation planning, and OBEXGATE deployment scoping.
Why it matters before enforcement.
Observer Mode shows how your system behaves. Enforce Mode determines what is allowed to run.
EVF comes before both. It identifies what must be understood, evidenced, and prioritised before runtime enforcement is switched on.
If you cannot identify the execution boundary, you cannot control it.
Who should use EVF.
EVF is for organisations that need a structured view of AI governance readiness before platform deployment, audit engagement, regulatory review, or board-level decision-making.
- Boards and executives needing a defensible governance picture
- Legal and compliance teams preparing for scrutiny
- CTOs and CISOs mapping implementation risk
- Public sector and regulated organisations with high-risk AI exposure
- Advisors and referral partners qualifying client readiness
A paid diagnostic line item.
EVF is available as a standalone paid diagnostic and as the intake layer for OBEXGATE deployment.
Pricing is provided after scoping, based on organisation size, number of systems, regulatory posture, and whether FRIA-oriented analysis is required.
Commercial terms are provided privately. EVF is not priced as a generic questionnaire.
Start with EVF.
Use EVF when you need something more concrete than a readiness score and more useful than a workshop transcript.
EVF is a diagnostic assessment. Information provided does not constitute legal advice, regulatory advice, financial advice, or a guarantee of compliance or outcome.