AI safety
New AI Safety Protections Announced. What They Mean at Work
On 10 November 2026, the Australian Government announced plans for mandatory safety and security protections for frontier AI companies that operate in Australia. The rules are planned and are not law yet. They are aimed at the companies that build the most advanced AI, so your workplace rules and your own habits still matter most today.
What did the Australian Government announce?
The announcement was a media release dated 10 November 2026. It says "strong mandatory safety and security protections will be legislated through the AI Standards". These would apply to frontier AI companies that want to operate in Australia.
The release also says the government will propose a "new targeted regulator for AI safety". It would work with the Australian AI Safety Institute. The release says the regulator could apply significant penalties to companies that do not comply.
It also says more detail will come once the government finishes its work and consultation. So the final shape of the rules is not settled.
Is this law yet?
No. The release describes protections that will be legislated. It does not say they are in force. The regulator is something the government will propose.
Until a law passes, treat this as an announcement of direction. Details could change.
What would frontier AI companies be required to do?
The release lists six things the protections would cover. Under the proposed approach, companies would be required to do the following.
- Maintain security controls across development and deployment.
- Identify and manage serious risks from their models.
- Allow independent evaluation of their models.
- Rapidly notify the government of incidents.
- Maintain and publish safety and security frameworks.
- Respond to updated standards and rules as the technology changes.
These are duties for the companies that build the AI. They are not duties for a clinic, council, school office or small business that uses an AI tool.
Does this change what I have to do at work?
The release does not describe new duties for workplaces that use AI. It is about frontier AI companies.
That does not mean you can relax. The Office of the Australian Information Commissioner (OAIC) says "The Privacy Act applies to all uses of AI involving personal information." Your organisation may also have its own AI policy and other rules for your role. If you are unsure which apply, ask your manager, privacy officer or legal contact.
Think of it this way. The planned rules aim to make the AI tools safer to build. They would not check the report you wrote with an AI tool this morning. That check is still your job.
What can you do now?
These are practical suggestions from AI Work Ready, not requirements in the announcement.
- Follow your AI policy. Find out which tools are approved and what you can use them for. If there is no policy, ask who is working on one.
- Check the output. Read what the AI gives you. Check names, numbers and claims before you send or act on them.
- Keep sensitive information out. Do not paste client details, personal information or confidential documents into a tool your organisation has not approved. The OAIC advises organisations not to enter personal information, and particularly sensitive information, into publicly available generative AI tools.
- Report problems inside your organisation. If an AI tool gives a harmful, wrong or strange result, or you think something sensitive went into the wrong tool, tell your manager or the person who handles IT or privacy. Do it early. A short note is enough.
- Know your tools. The National AI Centre's essential AI practices encourage organisations to understand what AI systems are operating within them and to have policies, guidelines and standards in place. Your team can start with a simple list.
Why does the incident notification item matter to you?
The release says frontier AI companies would need to rapidly notify the government of incidents. That is about the companies and the government. It does not replace reporting inside your own workplace.
If something goes wrong with AI at your job, your organisation still needs to hear about it from you. Use your internal process, and ask what it is if you do not know.
Where can you learn more?
If you want a practical starting point for using AI at work, the course Responsible AI for Australian Workplaces covers the basics. Teams can also look at AI Work Ready for organisations.
This is general information, not legal advice. Check your organisation's policy.
Sources
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.