Australia Proposes AI Safety Regulator And Mandatory Incident Reporting
On 10 November 2026, the Australian Government proposed mandatory safety duties for frontier AI companies and a targeted regulator. The proposal does not create an immediate compliance checklist for ordinary businesses, but it raises the questions buyers should ask AI suppliers.
On 10 November 2026, the Australian Government announced proposed mandatory safety and security duties for frontier AI companies operating in Australia. It also proposed a targeted AI safety regulator to enforce those duties with significant penalties.
This is a policy announcement, not a finished law. The government said further detail will come as it completes the legislation and consultation. Claims that Australia has already created criminal liability or a complete compliance regime go further than the official releases.
What The Government Announced
The proposed rules are aimed at frontier AI companies, meaning the developers of the most capable models. The government said those companies would be required to:
- maintain security controls during development and deployment;
- identify and manage serious model risks;
- allow independent model evaluations;
- notify the Australian Government rapidly about incidents;
- publish safety and security frameworks;
- respond when standards and rules change.
The proposed regulator would work with the Australian AI Safety Institute, which would provide technical expertise and oversight.
The government released the proposal alongside recommendations from a Rapid Review into an OpenAI agent incident involving the Medicare Statistics Reporting Service Portal. The review made 18 recommendations. The government said it accepted them and had started implementation work.
Those recommendations include mandatory reporting of misaligned behaviour or cyber incidents, stronger information sharing, and consideration of criminal law reform for serious misconduct. Consideration is the key word. The release did not create a new criminal offence on 10 November.
What It Means For An Australian Business Owner
The two releases do not set out new duties for an ordinary business that uses ChatGPT, Claude, Gemini or another AI product. The direct obligations are proposed for frontier model companies.
The useful change for a buyer is a clearer standard for supplier questions. If an AI provider has access to your email, customer records, files or business systems, you should be able to find out how it handles an incident and how quickly you will hear about one.
Do not wait for the final law to ask:
- What events does the provider call a safety or security incident?
- Will it tell customers when their data, account or connected system may be affected?
- Does an independent party test the model or service?
- Where is the current safety and security framework published?
- Who is responsible for acting on an alert inside your business?
A supplier may not have every final regulatory answer yet. It should still be able to explain its present controls and notification process in plain language.
The Step To Take First
Add four fields to your AI supplier register: incident contact, notification promise, published safety document and internal owner. Complete them first for tools connected to email, files, CRM, accounting or customer data.
If a field is blank, ask the supplier or implementation partner for a written answer. Save that answer with the system record. When the legislation is published, you will have one place to compare the final duties with the controls your providers already claim.
Our guide to the Australian Privacy Act and AI automation covers the existing privacy questions that still apply while the new AI rules are being developed.
Sources
Go Deeper
- Australian Privacy Act Compliance For AI Automations →
A plain-English guide to the Australian Privacy Act and AI automation: the framework that applies and the choices that keep you in control. Not legal advice.
- Is It Safe To Connect AI Automation To Your Business Data? →
How AI automation keeps your business data safe: built in your own stack, least-privilege access, human-in-the-loop, and what to ask before connecting anything.
Sam co-founded Echelon AI Solutions and leads transformation strategy, client engagements and growth. He has built and operated businesses across marketing and AI education, and has guided companies in retail, trades, hospitality and professional services through operational change. His focus is making AI earn its place through measurable business performance.