Veracity AI

Home
About
Services
Blog
Contact

AI Governance Readiness Diagnostic

10 minutes to identify where your Ai governance will hold — and where it won’t

This diagnostic assesses five dimensions of operational AI governance. Most organizations score well on policy and documentation. Most score poorly on authority, evidence, and sustainability. The gaps between these scores reveal where governance will fail under pressure.

Section 1: Decision Authority

1. If an AI system produced harmful outputs tomorrow, who has the authority to stop it before it affects customers?

A. A specific named individual with documented authority (3 points)

B. A role or function, but the specific person varies (2 points)

C. Everyone “knows” who would handle it, but it’s not documented (1 point)

D. Unclear or would require escalation to determine (0 points)

2. If that person exercised stop authority and it cost the company revenue, what would happen?

A. They would be protected — this is expected and supported (3 points)

B. It depends on whether they were right (2 points)

C. They would face questions or pressure (1 point)

D. It would be career-limiting (0 points)

Section 2: Evidence & Accountability

3. Can you produce, today, the evidence trail for your last 10 AI-influenced decisions?

A. Yes — captured automatically at the moment of decision (3 points)

B. Partially — some documentation exists but requires assembly (2 points)

C. We could reconstruct it, but it would take significant effort (1 point)

D. No — we don’t systematically capture this (0 points)

4. When an AI system’s output is overridden or rejected, is the reason captured?

A. Yes — automatically, with timestamp and rationale (3 points)

B. Sometimes — depends on the system and the person (2 points)

C. Rarely — overrides happen but aren’t tracked (1 point)

D. No — we don’t track overrides (0 points)

Section 3: Architecture & Controls

5. Are your AI governance controls bypassable?

A. No — controls are embedded in the system architecture (3 points)

B. Bypass requires explicit approval and is logged (2 points)

C. Technically possible to bypass, but against policy (1 point)

D. Controls are advisory — people can work around them (0 points)

6. If an AI system drifts outside its intended parameters, what happens?

A. Automatic alert and/or automatic pause until reviewed (3 points)

B. Monitoring would detect it, but response is manual (2 points)

C. We might notice eventually through downstream effects (1 point)

D. We don’t have systematic drift detection (0 points)

Section 4: Organizational Sustainability

7. When was the last time your governance framework was tested under real pressure (not audited — tested)?

A. Within the last 6 months — we run scenarios or had a real incident (3 points)

B. Within the last year (2 points)

C. More than a year ago or never formally tested (1 point)

D. We’ve never tested it — only audited or documented (0 points)

8. Do the people responsible for AI governance have the training and capability to exercise it?

A. Yes — explicit training, clear procedures, practiced regularly (3 points)

B. Some training exists, but it’s not comprehensive (2 points)

C. They’re expected to figure it out when needed (1 point)

D. Governance is assigned, not trained (0 points)

Section 5: Executive Alignment

9. Does AI governance have explicit C-suite sponsorship with authority to enforce?

A. Yes — named executive with budget, authority, and accountability (3 points)

B. Shared across multiple executives (2 points)

C. Delegated to a committee or working group (1 point)

D. No clear executive ownership (0 points)

10. When AI governance creates friction with business objectives, what happens?

A. Governance holds — friction is expected and managed (3 points)

B. Case-by-case negotiation (2 points)

C. Business objectives usually win (1 point)

D. Governance is quietly bypassed or ignored (0 points)

Please answer all questions before submitting.

Ready to strengthen your AI governance?

Book a consultation with Gary Trautmann, PhD to discuss your results and develop a plan to close your governance gaps.

Book Your Consultation

This diagnostic identifies where your governance is strong and where it’s vulnerable. The gaps between sections matter as much as the total score.

If you scored below 18, or if any single section scored below 3, those areas represent failure points under pressure.

I work with organizations to close these gaps — not with more frameworks, but with operational governance that holds when it matters.

If you’d like to discuss your results and explore whether there’s a fit, email me at gary.trautmann@ethicalveracity.ai

Gary Trautmann, PhD
AI Governance & Change Management | Founder, Veracity AI
gary.trautmann@ethicalveracity.ai | ethicalveracity.ai