Free — a working document, not an official submission

Assurance Pack Studio

A self-check against the National AI Centre's Guidance for AI Adoption and the WA Government AI Policy and Assurance Framework . This is a personal working document you bring to your own governance process — it does not submit anything, and it is not a compliance certification or legal advice.

Project name

1. Decide who is accountable

Someone senior owns AI governance, and there's a policy staff can actually read.

Assign a senior leader as the overall AI governance ownerWrite an AI policy that tells staff how to use AI responsibly and what the risks are

2. Understand impacts and plan accordingly

Know who this could affect, and give them a way to push back.

Carry out a stakeholder impact assessment, paying attention to vulnerable or marginalised groupsSet up a channel for people to report problems or challenge an AI-influenced decision

3. Measure and manage risks

Have a way to flag which AI uses need more scrutiny before they need it.

Create a risk screening process to flag AI uses that need extra governance attentionRe-assess risk when the use case, data, or deployment context changes

4. Share essential information

People should know when they're dealing with AI, and you should know what AI you're running.

Keep an AI register — what systems, what they're used for, whether built in-house or procuredDisclose to users when they're interacting with AI or AI-influenced content

5. Test and monitor

Check it works before it ships, and keep checking after it does.

Test the system against clear acceptance criteria before deploymentMonitor for behaviour changes or unintended consequences once it's live

6. Maintain human control

A person can always step in — this isn't optional once something looks wrong.

Define the point(s) in the workflow where a human can intervene or override the systemMake sure the accountable person actually has the access and authority to intervene

WA AI Advisory Board escalation check

The three real criteria the WA Government AI Policy uses to decide whether a project's self-assessment must go to the WA AI Advisory Board.

Residual risk (after mitigations) is mid-range or higherFunded through the Digital Capability FundTotal project cost exceeds A$5 million

Notes