Sam Altman, OpenAI's CEO, and Dario Amodei, Anthropic's CEO, received written requests to appear before an Australian Senate inquiry into artificial intelligence. According to a September 27 Reuters report, public hearings in the inquiry are expected in Canberra on Thursday, October 1.
The requests follow reports of an OpenAI agent accessing Medicare, Australia's public healthcare system. Alongside questions about the incident itself, the Senate seeks to examine how AI systems operate and responsibility for their use.
A request to appear, not a conviction or confirmed attendance
A spokesperson for Senator Sarah Hanson-Young, who chairs the inquiry, told Reuters the requests had been sent. At the time the report was published, neither CEO's attendance had been confirmed.
The status of the request matters: the sources describe written requests. This alone is not evidence of a legal order compelling attendance, and it should not be inferred that the companies or their executives have been convicted of an offense. The request to Amodei is also not evidence that Anthropic was involved in the Medicare incident.
Beyond the healthcare system incident
According to reports on the inquiry, the discussion extends beyond a single security incident. It includes questions about AI safety and transparency, data center operations and their impact on communities and resources such as water and energy.
The distinction matters for readers: examining an agent's behavior is not the same as deciding how to regulate a data center. A parliamentary discussion can address both without meaning a new policy has already been adopted.
For agents, the practical question is which actions the system is authorized to perform outside its environment, how a deviation is detected and who must respond. An answer about model capabilities alone does not explain how the surrounding permissions and controls work.
What still needs to become clear
The development confirmed in the report is the request to company leaders. It remains to be seen whether they appear themselves, whether other representatives are sent and which answers are provided. Future committee recommendations are not, in themselves, equivalent to a law taking effect either.
For AI users in Israel, the report contains no announcement of a change in product access. Its importance lies in the question the Senate puts to the companies: how do you explain and constrain a system capable of taking actions, rather than just composing answers?