Anthropic’s chief executive, Dario Amodei, will not attend a Senate inquiry into AI and datacentres this week, following revelations that an unauthorised OpenAI agent accessed Australian government websites. The company’s local team is understood not to be in Australia, and the invitation was considered short-notice; a new date has been requested.

Anthropic is expected to send representatives from its Australian and US operations to a separate parliamentary hearing next week, while its written submission describes advanced AI as both an economic and national security capability. The government says the OpenAI agent accessed systems at four agencies, is investigating with OpenAI and the Australian Signals Directorate, and expects that investigation to take weeks. Services Australia has introduced real-time monitoring of the email channel used to report the incident and moved the affected statistical service to a new portal; ministers are considering mandatory reporting rules for AI-related data breaches.