Notifications
Clear all
Topic starter
27/06/2026 1:29 pm
TL;DR: AI governance is measurable enough to support ownership, prioritisation, and control, as Collibra says its new Business Metrics dashboard turns AI activity into a quantified oversight view, using AI Trust Scores, risk exposure, and safeguards to help leaders see where trust is weak and action is needed, while Gartner predicts 60% of organisations will miss expected AI value by 2027 because governance is fragmented.
NHIMG editorial — what this means for NHI practitioners
By the numbers:
- Gartner predicts that by 2027, 60% of organizations will fail to realize the expected value of their AI initiatives due to fragmented governance frameworks.
- Trust is visualized as a heatmap across the AI lifecycle stages, banded from 0–100 up to 400+ assets.
Questions worth separating out
Q: How should organisations measure trust across AI use cases, agents, and models?
A: Organisations should use a single scoring model that compares AI use cases, agents, and models against the same governance criteria.
Q: When does AI risk reporting become useful for governance teams?
A: AI risk reporting becomes useful when it connects exposure to safeguards and ownership.
Q: What do security teams get wrong about AI oversight dashboards?
A: Teams often mistake visibility for control.
Practitioner guidance
- Define a common AI trust scoring model Set one scoring method for use cases, agents, and models so governance teams can compare assets using the same criteria.
- Map safeguards to the AI risk matrix Use a risk-versus-safeguards matrix to identify high-impact AI initiatives that have weak control coverage.
- Tie each score to a named control owner Assign responsibility for each AI asset so trust findings cannot sit in a dashboard without an accountable team.
What's in the full announcement
Collibra's full product update covers the operational detail this post intentionally leaves for the source:
- How the AI Trust Score is structured across use cases, agents, and models for dashboard use
- How the risk distribution view maps high, medium, and low risk against safeguard coverage
- How lifecycle-stage heatmapping is presented for intention, development, validation, deployment, innovation, and optimisation
- How product owners and governance teams are expected to use the dashboard in day-to-day oversight
👉 Read Collibra's update on AI Command Center business metrics →
AI trust scores and business metrics: can governance keep up?
Explore further
27/06/2026 2:51 pm
AI governance is becoming measurable, but measurability is not the same as control. Business metrics help leaders see where trust is weak, where safeguards are thin, and where ownership is unclear. That matters because most AI programmes fail first at governance coordination, not at model performance. The practitioner conclusion is simple: if oversight cannot be quantified, it cannot be governed with confidence.
A few things that frame the scale:
- 70% of organisations grant AI systems more access than they would give a human employee performing the exact same job, according to the 2026 Infrastructure Identity Survey.
- 52% of respondents see AI decision-making power shifting toward platform and infrastructure teams rather than the executive suite, according to the 2026 Infrastructure Identity Survey.
A question worth separating out:
Q: Who should own remediation when AI trust scores show weak controls?
A: The owner should be the team accountable for the AI asset and the safeguard gaps affecting it. Governance teams should not become the remediation team by default. Their role is to surface the issue, define the threshold for intervention, and make sure the accountable owner has to act.
👉 Read our full editorial: AI governance metrics expose where trust and risk are unmeasured
