AI governance & AI policy · New Zealand & Australia
AI governance and AI policy that people actually follow
Practical AI governance frameworks, policy documents and agent-powered oversight for NZ and Australian organisations. Manage risk, privacy and the impact on your people, while keeping AI adoption moving.
Prosci® Certified · Microsoft Copilot-certified · Independent of every AI vendor
Your free 30-minute governance check
A confidential conversation. You’ll leave knowing:
- Where AI is already being used, officially or not
- Your biggest gaps in policy, risk, privacy and people impact
- Which NZ and Australian obligations apply to you
- Three priority actions you can take straight away
No obligation. For leaders, boards, risk, HR and IT teams.
- 9 governance areas covered, including people and psychosocial risk
- 5 AI platforms supported: Copilot, Claude, Gemini, Grok and ChatGPT
- NZ & AU aligned to local guidance and obligations
Why AI governance matters now
AI governance has moved from good practice to expectation
Your people are already using AI. Regulators, boards and customers now expect you to show how it’s controlled. Here’s what has changed across New Zealand and Australia.
- Now
AI is already in use
Most organisations have staff using AI tools today, often without an agreed policy. That creates privacy, security and quality risks nobody is managing.
- Australia · Late 2025
Guidance for AI Adoption
The National AI Centre’s guidance replaced the Voluntary AI Safety Standard with six essential practices for responsible AI use.
- NSW · Passed Feb 2026
AI is now a WHS issue
New digital work systems duties require businesses to manage risks from AI and algorithms, such as excessive workloads, monitoring and biased decisions.
- Australia · 10 Dec 2026
Automated decision transparency
Privacy policies must explain where computer programs make or support decisions that significantly affect individuals.
- New Zealand · Now
Existing laws already apply
No AI-specific law, but privacy, consumer, employment, human rights and health and safety duties all apply to AI, guided by MBIE’s Responsible AI guidance.
General information, not legal advice. We work alongside your legal and privacy advisers where needed.
What good AI governance covers
Nine areas, one joined-up AI governance framework
Most AI policies stop at acceptable use. Ours cover the full picture, including the human side that most frameworks miss: psychosocial risk, workforce impact and change management.
-
Accountability & decision rights
Clear owners for AI, who approves new tools and use cases, and how decisions are escalated.
-
AI policy & acceptable use
Plain-English rules for what staff can and can’t do with AI, and which tools and data are approved.
-
AI register
A living record of every AI tool and use case across the organisation, with owners and risk ratings.
-
Risk & impact assessment
Assess legal, operational, ethical and reputational risk before AI is deployed, not after.
-
Privacy, data & security
Protect personal and confidential information, and meet privacy obligations including automated decision transparency.
-
Psychosocial & workforce impact
Manage how AI changes workloads, monitoring, job control and roles, so AI doesn’t create new health and safety risks.
-
Change management & adoption
Help people understand and follow the rules, so governance is used in practice, not filed away.
-
Vendor & third-party AI
Assess AI built into the software you buy, and set expectations for suppliers.
-
Monitoring, incidents & reporting
Review cycles, an incident process and clear reporting to executives and the board.
AI governance documents
Your AI policy and governance document pack
Every document is tailored to your organisation, sector, tools and obligations. No generic templates, and written in plain English so people can follow them.
Documents are aligned to Australia’s Guidance for AI Adoption and New Zealand’s Responsible AI guidance, and linked to your AI strategy so governance supports your priorities, not just your risks.
- AI policy
- Acceptable use policy and staff guidelines
- AI register (inventory of tools and use cases)
- Risk and impact assessment templates
- Psychosocial risk assessment for AI-enabled work
- Privacy and automated decision-making disclosures
- Vendor and AI tool assessment
- AI incident response plan
- Board and executive reporting pack
- Training and communication plan
Agent-powered AI governance
AI agents keep your governance current. People stay in control.
Governance fails when it becomes a document nobody updates. We set up AI agents that do the time-consuming work of keeping AI policy and oversight current, using the AI platform you already have. Learn more about our agentic AI implementation.
-
Drafts and updates policy
Prepares policy updates when your tools, use cases or obligations change.
-
Tracks regulatory change
Monitors NZ and Australian guidance and flags what affects you.
-
Maintains the AI register
Keeps a current record of AI tools, use cases, owners and risk ratings.
-
Runs approval workflows
Routes requests for new AI tools and use cases to the right approver.
-
Answers staff questions
Gives staff instant, consistent answers about what the policy allows.
-
Produces board reports
Summarises AI use, risks, incidents and actions for executives and the board.
Named owners review and approve every policy change and every AI decision that matters. That’s what “maintain human control” means in practice.
- Works with:
- Microsoft Copilot
- Claude
- Gemini
- Grok
- ChatGPT
Aligned to NZ and Australian guidance
How we deliver Australia’s six essential practices
The National AI Centre’s Guidance for AI Adoption sets out six practices for responsible AI. Here’s how each one is built into your framework.
- 1
Decide who is accountableGovernance structure, AI owners, decision rights and approval workflows.
- 2
Understand impacts and plan accordinglyImpact assessments covering customers, staff, psychosocial risk and workforce change.
- 3
Measure and manage risksAI risk register, risk ratings and vendor and tool assessments.
- 4
Share essential informationAI register, staff guidance, and privacy and automated decision disclosures.
- 5
Test and monitorAgent-supported monitoring, review cycles and an incident process.
- 6
Maintain human controlHuman review points, named approvers and clear escalation for every AI decision that matters.
In New Zealand? The same framework covers MBIE’s Responsible AI guidance for businesses and your obligations under the Privacy Act 2020 and the Health and Safety at Work Act 2015, which covers mental as well as physical health.
Governance people follow
A policy nobody follows isn’t governance
The biggest AI governance risk isn’t a missing document. It’s staff who don’t understand the rules, don’t see why they matter, or quietly work around them.
That’s why every engagement includes structured AI change management based on the Prosci ADKAR® model, plus practical AI training and executive AI coaching so leaders can role-model safe, confident AI use.
- A Awareness
- D Desire
- K Knowledge
- A Ability
- R Reinforcement
Independent, people-first governance
- Vendor-neutral : no software to sell, no commissions
- Prosci® Certified Change Practitioner
- Microsoft Copilot-certified ( verify)
- 25+ years leading digital transformation and change
Verified Prosci® certification. See the Credly badge and more about Matrix AI.
How it works
From first conversation to governance that runs
-
Free 30-minute check
A confidential look at how AI is used today, your main gaps and three priority actions.
-
Review and gap analysis
We map your AI use, risks and obligations, including privacy, psychosocial and workforce impact.
-
Build and set up
Tailored policy documents, your AI register and agent-powered oversight on your AI platform.
-
Embed and review
Training, communication and regular reviews so governance keeps pace as AI use grows.
Every organisation we work with gets AI governance support, including construction, engineering and land development clients across New Zealand and Australia.
FAQs
AI governance and AI policy questions
What is an AI governance framework and why does it matter?
An AI governance framework sets out how AI is approved, used, monitored and held accountable across your organisation: who owns it, which tools and uses are allowed, how risks are assessed and how issues are escalated. Without one, AI use becomes fragmented and hard to defend when something goes wrong.
What is the difference between AI policy and AI governance?
An AI policy is the set of rules: what staff can and can’t do with AI, and which tools and data are approved. AI governance is the system that makes those rules work: ownership, decision rights, risk assessment, the AI register, monitoring, training and reporting. You need both.
Do NZ and Australian organisations legally need AI governance?
There is no single AI law in either country, but existing laws already apply to AI, including privacy, consumer, employment, human rights and health and safety law. In Australia, new automated decision transparency obligations start on 10 December 2026, and NSW has passed digital work systems duties covering AI. Governance is how you show you are managing these obligations. This is general information, not legal advice.
How do AI agents help manage AI governance?
Agents do the repetitive work: drafting policy updates, tracking regulatory change, keeping the AI register current, routing approval requests, answering staff questions and preparing board reports. They work with Microsoft Copilot, Claude, Gemini, Grok or ChatGPT. People stay in control: named owners review and approve every policy change and every decision that matters.
What happens in the free 30-minute AI governance check, and what does governance cost?
In 30 minutes we look at how AI is being used in your organisation, your main gaps and the obligations that apply to you, and agree three priority actions. It is free and confidential. Governance work is priced on application, in AUD or NZD, with a fixed quote once the scope is agreed.
Book your free 30-minute AI governance check
Tell us your organisation, the AI tools your people use and your biggest governance concern. We’ll come back with a time for your free, confidential 30-minute check. Prefer email? hello@matrixconsulting.ai