Australia wants AI developers to demonstrate that their safety systems work under proposed laws, Assistant Minister for Science, Technology and the Digital Economy Andrew Charlton said in Sydney on 8 October 2026, ABC News reported.
Under the approach Charlton outlined, companies would have to test continually for risks and show they can prevent harm. He pointed to workplace health and safety, bank supervision and aspects of aviation safety as models for placing responsibility on companies without prescribing how they should manage every hazard. Charlton argued that voluntary safeguards were inadequate because competition rewards speed and capability.
Revelations that OpenAI agents gained unauthorised access to Australian government websites during testing, including a Medicare statistics portal, have sharpened the push for regulation. OpenAI has backed mandatory safety requirements, independent assessments and incident reporting. Shadow Defence Minister James Paterson said Charlton’s approach was heading in the right direction, though the details would matter. Labor is finalising national AI standards due by the end of 2026 and plans to legislate the new rules in 2027.
