AI readiness
Six Essential AI Practices for Your Australian Workplace
The Australian Government's Guidance for AI Adoption sets out six essential practices for using AI safely and responsibly. They are decide who is accountable, understand impacts and plan accordingly, measure and manage risks, share essential information, test and monitor, and maintain human control. This article walks through each one and gives you a small step to try this week.
Where does this guidance come from?
The Guidance for AI Adoption was published by the Department of Industry, Science and Resources on 21 October 2025. The National AI Centre now hosts the current guidance on AI.gov.au. That site brings together practical guidance, tools and resources from across government in one place. Its first release focuses on small and medium businesses and the not-for-profit sector.
There are two versions. Foundations is for organisations starting out, using AI in low-risk ways, or new to AI governance. The Implementation practices version suits organisations with mature governance, technical teams or higher-risk uses. This article follows Foundations.
You should not treat it as a checklist to finish in a week. The guidance itself says, "You don't need to do everything at once."
What does "decide who is accountable" mean?
Your organisation stays responsible for how AI is used. The guidance suggests naming a senior leader as the AI governance owner and having an AI policy.
This week, your team could agree on one person who answers questions about AI use. A council records team might put that name at the top of a one-page note. An aged care provider might ask the facility manager to hold that role for now.
How do you understand the impact on people?
AI works at speed and scale, so problems such as unfair decisions or wrong information can spread widely. The guidance asks you to think about who is affected, especially vulnerable groups, and to give people a way to raise a concern.
This week, pick one AI tool you use and list who it touches. An early childhood service might note that a tool drafting newsletters reaches every family. Then write down where a parent can say something looks wrong.
How do you measure and manage AI risks?
The guidance says risk depends on how a tool is used, not only on the software. Drafting a staff meeting agenda is different from summarising a patient record.
This week, sort your AI uses into low, medium and high. A medical practice might mark anything near patient information as high, and pause it until the right people have looked at it.
What essential information should you share?
People should know when they are dealing with AI, and when AI affects decisions about them. The guidance points to keeping an AI register and telling people when you use AI.
This week, start a simple list with three columns for the tool, what it is used for, and who looks after it. An office team can build this in a shared spreadsheet in about ten minutes. You should also decide where you will tell clients or the public about your AI use.
How do you test and monitor AI tools?
AI behaviour can change over time and can be less predictable than ordinary software. The guidance suggests asking suppliers for proof of testing, testing before you roll a tool out, and checking it once it is running.
This week, take one task and run it through your AI tool with five examples where you already know the right answer. A council customer service team could use past, de-identified enquiries. Note anything the tool gets wrong. Keep personal information out of the tool unless your policy allows it.
What does "maintain human control" look like day to day?
AI tools need human oversight that matches the stakes. The guidance says to build in points where a person can pause, override or shut a system down, and to keep another way of doing critical work.
This week, choose one AI-assisted task and write down who checks the output before it goes out. In aged care, that might be a care manager reviewing any AI-drafted summary before it is filed. Also note what you would do if the tool stopped working tomorrow.
What should you do next?
Start with one practice. The guidance says to "Adapt each practice to fit your organisation's size, use cases and risk profile." Small steps count, and they give your team something real to build on.
If you want help working out where your team sits, try the AI Readiness Check or read about our work with organisations.
This is general information, not legal advice. Check your organisation's policy.
Sources
- Guidance for AI adoption, Department of Industry, Science and Resources, 21 October 2025
- Guidance for AI adoption: foundations, Department of Industry, Science and Resources
- Guidance for AI adoption: foundations, AI.gov.au, National AI Centre
- National AI Centre launches AI.gov.au, Department of Industry, Science and Resources
Learn the habit, free
Responsible AI for Australian Workplaces teaches what never goes into an AI tool, how to ask well and how to check the answer. Short online lessons, free for individuals.