HAOP Diagnostic

True Function Diagnostic

Most AI-enabled workflows can show that they manage safety. Fewer can show that they produce it. This tool runs the True Function Test against one workflow you choose, forces the answers across the human, AI, and organizational performers, then grades each control by how far it has moved from claimed to owned.

It does not give you a score to put on a slide. It gives you a state, the gaps that must close before the workflow is trusted with consequential work, and the first move to make.

Name one workflow What does it do, in one or two lines?

Everything stays in your browser. Nothing is sent anywhere.

1 · Diagnostic

2 · Alignment read

3 · Result

Stage 1

Run it across the three performers

Answer what is true, not what the policy says. Empty cells are findings too. Answered 0 / 30 cells

Red-flag review

Detected patterns are pre-checked. Confirm or clear each against what you actually know about the workflow.

1 · Diagnostic

2 · Alignment read

3 · Result

Stage 2

Grade each control by evidence

Grade the evidence you could show an auditor, not your confidence. A control that lives only in a policy is a 1, not a 3. Controls marked critical are non-negotiable: a 0 on any one of them holds the whole workflow at Representational Risk regardless of the rest.

0 · Not defined

The control does not exist for this workflow.

1 · Claimed

Asserted in policy or intent, not verified in practice.

2 · Partial

Verified somewhere, inconsistent or dependent on individuals.

3 · Owned

Designed, verified, and owned. Survives the person who built it.

1 · Diagnostic

2 · Alignment read

3 · Result

Alignment heat map

Hard-stop gaps

The True Function Diagnostic does not certify a workflow as safe and is not a maturity score. It measures whether each control that produces the outcome is real or only represented, gates the criticals, and points to the first move. Re-run it as the work changes.