Free tool

Free AI register template

Record every AI system and use case in your organisation, screen each one with the Australian Government's seven screening questions, and export the register to a spreadsheet or Word document.

Runs in your browser. Nothing you enter is sent to AusGPT.

Register details

Optional. These appear at the top of your Word and Markdown exports.

The person who keeps the register up to date and is the contact point for it.

Your AI register

Saved in this browser only. Download a copy to share it or keep it somewhere central, such as your intranet or a shared drive.

AI system and use Owner Status Screening outcome Personal information Next review
Add an AI system

Add an AI system or use case

Fields marked * are required. Fields follow the National AI Centre's AI register template, plus personal information, data location, human oversight and review date.

1. The system

A clear name for the AI system or use, with a version where needed. Record embedded AI features too, such as AI in your CRM or meeting software.

The person or role accountable for this system.

Who built or provides it, for example an internal team or a named vendor.

2. Purpose and use

Why the system exists and what it aims to achieve.

The specific ways it is meant to be used.

What the system can't or shouldn't do, and what staff must not use it for.

Ways the system could be misused, or "None identified".

3. Data and people

What data it learns from or processes, and its type (structured data, text, images, audio).

Who might be affected, for example customers, staff, job applicants or the public.

For example "Australia only (provider confirms in contract)" or "Unknown: ask provider".

4. Risk screening

The seven AI screening questions from the National AI Centre. Answering yes to any question means this use may need higher levels of governance oversight.

1. Does the AI system access, process or handle any personal, sensitive or confidential information during training, operation or as input?

For example customer personal data, employee information, health records, financial data or confidential company information used as input or in prompts.

2. Does the AI system interact directly with end users with significant autonomy or freedom to respond without meaningful human oversight?

For example LLM-based chatbots that generate varied responses to customer queries, or assistants that give advice beyond pre-defined scripts.

3. Does the AI system act autonomously at scale, or would it be difficult to maintain human control and intervene if needed?

For example AI agents that can take actions across multiple systems or have internet access beyond low-risk, constrained tasks.

4. Does the AI system make or support decisions affecting people in vulnerable circumstances, marginalised groups or potentially underrepresented populations?

For example children, older people, people living with disability, Indigenous communities or low-income groups.

5. Does the AI system operate in a regulated area, or can it have legal effects on individuals?

For example financial approvals, healthcare recommendations or eligibility determinations.

6. Would harm from a wrong decision be difficult to contest, reverse or provide redress for?

For example automated communications that damage relationships, decisions affecting reputation or opportunities, or irreversible actions.

7. Is the AI system designed for multiple purposes, or easily adapted beyond its intended use through interfaces that are not tightly controlled?

For example systems built on general-purpose large language models, or tools that accept open-ended natural language instructions.

Answer the questions above, then choose an outcome. The screening tool gives no automatic score: the outcome is your organisation's judgement.

5. Oversight and review

Who checks the output, when a person must review or approve it, and who can pause or switch the system off.

When the system was first added to the register.

For example links to the risk assessment or privacy impact assessment, contract dates or approvals.

This free AI register template turns the Australian Government's register and screening templates into a working register you can fill in on this page. Add each AI system or use case, answer the seven screening questions, record who is accountable and when it's due for review, then download the whole register as a CSV for Excel or Google Sheets, or as a Word document to circulate.

What the AI register template records

The National AI Centre's Guidance for AI Adoption (October 2025) sets out six essential practices. Under practice 4, "Share essential information", the first action is to create and maintain an AI register that documents all your AI systems, how you use them and other important information. That includes AI you built, AI you bought and AI embedded in other systems, such as HR or customer engagement tools.

The register on this page uses every field in the government's template (Word and Excel versions published on ai.gov.au in April 2026) and adds four that most organisations end up needing:

Field What to record Source
System name and version A clear name for the system or use, with a version where it matters Government template
System owner The person or role accountable for the system Government template
Status Proposed, pilot, operational or decommissioned Government template (Proposed added)
Source Who developed or provides it Government template
Purpose / business goals Why it exists and what it should achieve Government template
Intended use case(s) The specific ways it is meant to be used Government template
Known limitations and prohibited use What it can't or mustn't be used for Government template
Foreseeable misuse How it could be misused Government template
Data sources and type What data it learns from or processes Government template
Key stakeholders affected Who could be affected Government template
Registered date When it was added to the register Government template
Screening outcome Normal, Elevated or Unacceptable Government template
Personal information involved None, personal or sensitive information Added
Where data is processed and stored Australia, overseas or unknown Added
Human oversight Who checks outputs and who can switch it off Added
Next review date When the entry must be checked again Added

The template is a starting point. The government's own instructions say to add fields you need for compliance, remove ones that don't add value and use your organisation's terminology.

How to screen each AI use case

The screening section uses the National AI Centre's seven AI screening questions. They ask whether the system:

  1. handles personal, sensitive or confidential information
  2. interacts directly with users without meaningful human oversight
  3. acts autonomously at scale, or would be hard to control
  4. makes or supports decisions affecting people in vulnerable circumstances or marginalised groups
  5. operates in a regulated area or can have legal effects on individuals
  6. could cause harm that is hard to contest, reverse or put right
  7. is general-purpose or easily adapted beyond its intended use.

The questionnaire has no scores or automatic outcomes. If you answer yes to any question, the use may need higher levels of governance oversight, and the guidance recommends consulting senior leadership or a governance committee before detailed planning. If every answer is no, continue with your standard AI adoption process. The tool shows that guidance as you answer, but you choose the outcome.

A general-purpose generative AI assistant will usually get a yes to question 7, because the screening tool itself gives general-purpose large language models as an example, and often to question 1, because staff put personal or confidential information into prompts. That doesn't make it unacceptable. It means the use needs clear rules, an accountable owner and checks that match the risk.

Tips for an AI register people actually use

  • Appoint a register administrator. The government template recommends one person who maintains the register, keeps a record of changes and is the contact point for it.
  • Log systems when they're approved. Anyone procuring or managing a new AI system should add it once it is approved, and update it throughout its use. Use the "Proposed" status to record screening before approval.
  • Include embedded and free tools. AI features switched on inside software you already use belong on the register, and so do free tools staff use for work. The National AI Centre notes a single register helps reduce "shadow" AI use.
  • Record where data goes. The OAIC's guidance on commercially available AI products asks organisations to consider whether personal information could be disclosed outside Australia. Recording the processing location now saves a scramble during a privacy review. Our security page shows the kind of detail to ask providers for.
  • Set review dates. Review an entry when the system changes, when the way you use it changes, and at a regular interval. Rows past their review date are flagged.
  • Keep the master copy somewhere central. The register on this page is saved only in your browser. Download it and keep the official version on your intranet or a shared drive.

What to do after you've built the register

A register shows what you have. It doesn't assess it. For any use that involves personal information, run a privacy impact assessment before you rely on it. To see how your wider AI governance measures up against the six essential practices, take the AI readiness assessment.

Commonwealth agencies should check the Digital Transformation Agency's Standard for accountability, which sets the minimum fields their internal AI use case registers must capture, and add any fields this template doesn't include.

Sources

  1. National AI Centre: AI systems register (template in Word and Excel)
  2. National AI Centre: AI screening questions
  3. National AI Centre: Guidance for AI adoption: foundations
  4. Digital Transformation Agency: Policy for the responsible use of AI in government, version 2.0 (strategy and oversight)
  5. OAIC: Guidance on privacy and the use of commercially available AI products

Frequently asked questions

What is an AI register?
An AI register (also called an AI inventory or AI use case register) is a table or database that lists every AI system your organisation uses, with its key characteristics, use cases, accountable people and the level of governance it needs. The National AI Centre recommends keeping one as part of the Guidance for AI Adoption.
Do Australian organisations have to keep an AI register?
For most businesses it isn't a legal requirement. The Guidance for AI Adoption recommends it as a 'getting started' action. Non-corporate Commonwealth entities are different: the Digital Transformation Agency's Policy for the responsible use of AI in government (version 2.0, effective 15 December 2025) requires them to create a register of in-scope AI use cases within 12 months, using the minimum fields in the DTA's Standard for accountability.
Should free tools like ChatGPT and AI features inside other software be on the register?
Yes. The National AI Centre says the register should capture all AI systems, including AI features embedded in common software packages. Recording the tools staff actually use, including free ones, is also one of the simplest ways to bring 'shadow' AI use into view.
What do Normal, Elevated and Unacceptable mean?
They are the screening outcome values in the government's register template. The screening questions themselves give no score: if you answer yes to any question, the use may need higher levels of governance oversight, and you decide which outcome fits after consulting the right people in your organisation.
Is an AI register the same as a risk assessment?
No. The National AI Centre describes the register as a transparency and documentation tool that should not replace more in-depth governance activities such as impact assessments and risk management. Use the register to see what you have and decide where deeper assessment is needed.
Where is my register stored?
Only in your web browser's local storage on this device. Nothing is sent to AusGPT. Clearing your browser data deletes it, so download a CSV or Word copy and keep the master version somewhere your team can find it.

Keep reading

AI governance AI readiness assessment Fifteen questions that score your organisation against the six essential practices in the Australian Government's Guidance for AI Adoption, with a shadow AI risk indicator and specific next steps for every gap. Privacy AI privacy impact assessment A guided privacy impact assessment for a proposed generative AI use. Describe the project, map the information flows, answer questions on the Australian Privacy Principles and get a structured draft PIA report with risks and recommendations. AI governance AI transparency statement generator Draft an AI transparency statement in the structure the Digital Transformation Agency requires. Built for Commonwealth agencies, and easy to adapt for councils, universities and businesses that want to publish one voluntarily. AI governance Guidance for AI Adoption (AI6) The National AI Centre's Guidance for AI Adoption is the government's main playbook for using AI responsibly. It updated the Voluntary AI Safety Standard in October 2025. Here are the six practices, the free templates and a 90-day plan for a small or mid-sized organisation. Security How AusGPT keeps data in Australia Where your data is stored and processed, and the controls that protect it.

Secure AI for your team, processed in Australia

Claude AI, document chat and voice dictation, with your data stored and processed in Australia. $29 per user per month.

Start free trial Talk to us