Broad, all roles
AI governance and adoption
7 min read
Australia's Guidance for AI Adoption: the six essential practices
GP
Guru Prasad, Founder, AAAI. Published 12 September 2026.
The Guidance for AI Adoption is a set of six essential practices released by the National AI Centre on 21 October 2025: decide who is accountable, understand impacts and plan accordingly, measure and manage risks, share essential information, test and monitor, and maintain human control. It builds on Australia's 2019 AI Ethics Principles and evolves the Voluntary AI Safety Standard's ten guardrails from August 2024 into a shorter, more practical set, without replacing either. It remains voluntary, but it is quickly becoming the checklist regulators, auditors and enterprise clients expect an organisation using AI to already be working through.
In brief
- The National AI Centre released the Guidance for AI Adoption on 21 October 2025, condensing the 2024 Voluntary AI Safety Standard's ten guardrails into six essential practices.
- It builds on Australia's AI Ethics Principles (2019) rather than replacing them, and it is a separate thread from the mandatory AI Standards process still being designed.
- Organisations that already built governance around the original ten guardrails do not need to start again: the six practices map directly onto that earlier work.
Plain English
"Evolves" here means the ten guardrails have not been withdrawn. The National AI Centre has effectively relabelled and regrouped them into six practices that are easier to action, with a companion crosswalk showing which guardrail sits under which practice.
This sits alongside, not on top of, Australia's AI Ethics Principles, covered in our explainer on the AI Ethics Principles. The principles describe what responsible AI looks like; the six practices are the practical checklist for getting there.
Australian context
The Guidance for AI Adoption was published by the National AI Centre, part of the Department of Industry, Science and Resources, on 21 October 2025. It sets out six essential practices: decide who is accountable, understand impacts and plan accordingly, measure and manage risks, share essential information, test and monitor, and maintain human control.
The Department describes the guidance as evolving the Voluntary AI Safety Standard released in August 2024, consolidating that standard's ten guardrails into the six practices above rather than replacing them. It builds on, and does not replace, Australia's AI Ethics Principles published in 2019. It is a separate development from the mandatory AI Standards process, which was still being designed as of this writing.
Sources: National AI Centre, Guidance for AI Adoption, October 2025; Department of Industry, Science and Resources, Australia's AI Ethics Principles, 2019; Voluntary AI Safety Standard, August 2024.
What changed, and why it matters
Ten guardrails asked a lot of organisations without always telling them where to start. Feedback on the Voluntary AI Safety Standard was that it was broad and principles-based in ways that made practical implementation genuinely uncertain, so the National AI Centre rebuilt it as six essential practices with clearer, more operational guidance and a crosswalk back to the original ten. The risk in ignoring this update is not a fine: neither the guardrails nor the six practices are mandatory. The risk is falling behind the reference point that regulators, auditors and enterprise procurement teams increasingly use to judge whether an organisation's AI use is responsible.
The opportunity runs the other way. If your organisation already mapped its AI governance to the ten guardrails, that work is not wasted: it maps directly onto the six practices. And if you have not started, six practices is a considerably shorter list to work through than ten, and a more practical head start on the mandatory AI Standards still taking shape through 2026 and 2027.
The six essential practices
Decide who is accountableName a person accountable for each AI system, including where a vendor or contractor supplies the tool. Accountability cannot sit with "the software".
Understand impacts and plan accordinglyIdentify who could be affected by an AI system, privacy, safety, fairness and security, before it goes live, not after something goes wrong.
Measure and manage risksFold AI-specific risks into your existing risk register rather than treating AI governance as a separate, parallel process.
Share essential informationTell affected people when AI is being used and keep an internal record of what each system does, what data it uses and how it is controlled.
Test and monitorCheck a system against accuracy and bias criteria before launch, then keep monitoring its performance, not just its uptime.
Maintain human controlKeep a person able to intervene, override or shut down an AI system, with a clear escalation path for high-stakes decisions.
What this means for you
Practitioners: HR, finance and consulting professionals working with AI day to day
You are usually the first to notice when an AI tool's output looks wrong for a particular person, which is exactly what "understand impacts" and "test and monitor" ask someone to be watching for.
Do this: for the AI tools you use most, check you know who is named as accountable for them, and where you would raise a concern.
Directors and delivery leads
Six practices are a shorter list to build a team process around than ten guardrails were, and a natural structure for a recurring governance review.
Do this: use the six practices as the agenda for your next AI governance check-in, rather than reviewing tools ad hoc.
Executives and boards
Being able to show alignment with the Guidance for AI Adoption is becoming a practical differentiator in government and enterprise procurement, and a head start on the AI Standards still being designed.
Do this: ask for a one page summary of which of the six practices are already in place across the organisation, and which are not.
Frequently asked
Is the Guidance for AI Adoption mandatory?
No. Like the AI Ethics Principles and the Voluntary AI Safety Standard before it, the Guidance for AI Adoption is voluntary. There is no fine for ignoring it, but it is the current benchmark government, regulators and enterprise clients use to judge whether an organisation's AI use is responsible, and it is a practical head start on the mandatory AI Standards still being designed.
How does the Guidance for AI Adoption relate to Australia's AI Ethics Principles?
The eight AI Ethics Principles, published in 2019, set out what responsible AI looks like at a high level. The Guidance for AI Adoption, released by the National AI Centre in October 2025, is a practical layer of six essential practices that build on those principles without replacing them.
What happened to the Voluntary AI Safety Standard's ten guardrails?
They have not been withdrawn. The National AI Centre describes the Guidance for AI Adoption as evolving the August 2024 Voluntary AI Safety Standard, consolidating its ten guardrails into six essential practices with clearer, more practical implementation guidance. Organisations that built governance around the original ten guardrails do not need to start again.
The six essential practices are a working checklist, not a certificate. If you want a structured way to build this capability across your organisation, AI Core is the place to start. If you are the person expected to own AI governance and strategy formally, CAIS: Certified AI Strategist is built around exactly this kind of judgement. For the full picture of how this fits with Australia's other AI rules, see our Australian AI regulation overview.
Explore AI Core
Explore CAIS
