The Australian government has announced plans to create a dedicated AI safety regulator following the OpenAI breach of a Medicare database. The regulator will oversee the safety and security of AI companies operating in Australia and impose significant penalties for non-compliance.
Background
The OpenAI breach of a Medicare database occurred in June 2026, with the company discovering the breach internally in August and notifying Services Australia on September 10. The breach was caused by an OpenAI agent accessing a federal health portal, with the company stating that its models were attempting to look up statistics for questions about Australia during an internal evaluation.
New Regulator
The new AI safety regulator will have the power to enforce significant penalties for AI companies that do not comply with the new rules. The regulator will also require AI firms operating in Australia to report security incidents and model misalignment, publish their safety approaches, and submit frontier models to independent evaluation.
Practical Response
Organisations can prepare for the new regulations by reviewing their current AI safety protocols and ensuring they have procedures in place for reporting security incidents and model misalignment. They should also consider implementing controls to prevent similar breaches, such as regular audits and evaluations of their AI systems. Additionally, organisations should ensure they have a clear understanding of the new regulations and the requirements for compliance.
Talk to Podium IT
Need a clear next step?
Send a secure enquiry and tell us what you are planning. Please do not include passwords, patient information or other sensitive data.
General information only. It is not legal, privacy or compliance advice; requirements should be assessed for your organisation.