Outcome

Compliance dashboard (ISO 42001)

Track Trust Score metrics, map controls, and demonstrate compliance with the ISO 42001 AI management standard.

Auditable AI is table stakes for regulated industries. Your board, your regulators, and your customers all want the same thing: evidence that the numbers on the screen are trustworthy and that a human is still in charge. The Compliance dashboard shows you exactly where you stand against the ISO/IEC 42001 AI management standard, so you never walk into an audit guessing.

Screenshot of the ISO 42001 compliance dashboard with trust score and control coverage

Open the Compliance tab

Open any command center and select the Compliance tab from the left-hand navigation. The tab is available to owners and admins by default, and you can grant external reviewers the Auditor role from the Permissions tab when you want them to see the same view you do.

The tab loads a single page focused on one question: how auditable is this command center right now? Everything on the page is computed live from the data already flowing through Delphi, so there is nothing to configure and nothing to schedule.

Trust Score metrics

At the top of the page, your Trust Score appears as a percentage ring. The score reflects how many of your KPIs have complete lineage graphs attached, so every headline number on your dashboard can be traced back to a source.

Three supporting gauges break down the score further. Provenance Coverage shows the share of datasets with an Aegis cryptographic proof attached. Ledger Integrity reports how many mutation events have been captured in the tamper-evident ledger. Actor Attribution shows the share of notifications and alerts that can be linked back to a named human or agent actor.

If any of your KPIs are missing lineage, Delphi lists them in a KPI Lineage Gaps callout directly below the Trust Score. Each gap is a concrete next action. Fix the gaps and your score climbs.

ISO 42001 control mapping

Further down the page, Delphi maps the live state of your command center against ten controls from ISO/IEC 42001, the international AI management standard. Covered controls include AI system lifecycle processes, impact assessment, leadership and governance, risk assessment and treatment, risk monitoring, human oversight mechanisms, data governance and quality, data provenance and lineage, performance monitoring, and continual improvement.

Each control is labelled covered, partial, or gap based on the same signals that drive the Trust Score. Partial and gap rows tell you where to focus next. Covered rows give you something concrete to point at when a reviewer asks how a given control is satisfied in practice.

Showing it to auditors

When an audit comes around, invite your reviewer to the command center as an auditor. The auditor role is read-only, scoped to the dashboards you explicitly share, and designed for exactly this situation.

From there, walk the auditor through the Compliance tab first, then the Audit tab so they can see the underlying ledger events that back every metric. The Compliance tab is the executive summary. The Audit tab is the receipts. Together they let you answer almost any “show me the evidence” question in minutes rather than weeks.

Delphi does not issue ISO 42001 certifications, and the dashboard is not a substitute for a formal audit. What it does do is give you and your auditors a shared, live view of the signals a certifying body will ask about, so you walk into the conversation prepared.

Built to deliver on better outcomes.

Schedule a demo or contact us to learn more.