Why Australia is scrambling over Anthropic's Mythos AI model

Australia's Home Affairs ministry and central banks move fast as Mythos exposes flaws

Anthropic's new Mythos AI model has done something no security audit ever managed at scale: it uncovered thousands of major vulnerabilities across every major operating system and web browser during a limited preview release. Now governments are scrambling.

A spokesperson for Australian Home Affairs Minister Tony Burke confirmed on Thursday that the government is actively working with software providers and Anthropic directly to track emerging vulnerabilities exposed by Mythos.

The statement marks one of the first official government acknowledgements of the model's potential to disrupt the established security landscape. Mythos was built for defensive cybersecurity purposes, but its autonomous capabilities have blurred the line between shield and weapon.

The Reserve Banks of Australia and New Zealand have reported in a joint statement today that they are keeping an eye on the deployment of Mythos and have been in touch with leading regulators across the world.

Both banking sectors are dependent on ageing IT infrastructure, with some being decades old, putting them at risk for attacks facilitated by Mythos' capabilities.

Australian Banking Association CEO Birmingham Simon, a body representing all commercial banking institutions in Australia, has stated that the banking sector is in discussions with the relevant regulators to safeguard the financial system.

The concern raised by cybersecurity professionals about Mythos is that due to its code structure and reasoning abilities, the time frame available between identifying an exploitable vulnerability and the attack itself can be significantly reduced.