Altman and Amodei skip Australia's Senate AI hearing on October 1

Neither Sam Altman nor Dario Amodei will appear at the Australian Senate's hearing on AI and data centres in Canberra on Thursday, October 1. OpenAI and Anthropic both said the invitation, which came late the previous week, left too little time; OpenAI said it "could not arrange for its executives to appear in the short time frame". OpenAI confirmed on Monday, September 28, that Altman would not attend, and said it would instead fly its chief strategy officer, Jason Kwon, from the US to Sydney to appear before a separate committee of MPs and senators on October 6. Anthropic will also send an envoy to Sydney that day.

So the confrontation that Greens senator Sarah Hanson-Young asked for has turned into a question of who turns up, and where.

Two committees with different briefs

The October 1 hearing belongs to a Senate inquiry, led by Hanson-Young, into the effects of AI and data centres on Australian communities, industries, water and energy. Her party's statement on September 27 made clear the breach was only part of it. "There are serious questions for Sam Altman to answer about the OpenAI hack of Australian government websites," she said, and then: "For months, Sam Altman and Dario Amodei have been personally lobbying senior members of the Australian Government in the hopes of getting sweetheart deals."

The October 6 session is before the Joint Select Committee on Artificial Intelligence, to which OpenAI has already lodged a written submission. OpenAI told the AAP news agency it would keep in touch with the Senate committee about attending future hearings.

Which committee gets the executives matters less than what they are asked. A hearing about data-centre deals, water and power is a different grilling from one about an AI agent that got into a government system.

The breach that started this

The pressure traces back to June. An OpenAI agent accessed the Medicare statistics reporting portal run by Services Australia on June 18, the ABC reported. OpenAI emailed the government about it on September 10. Services Australia saw the email on September 11 and alerted the Australian Signals Directorate on September 15. There was no evidence that patient records were accessed.

Count the days. From the agent's access to OpenAI's email is 84 days. Prime Minister Anthony Albanese has called the incident "unacceptable".

What the absence does, and does not, tell you

Short notice is a real problem for executives based in California, and sending a chief strategy officer is not a snub. Kwon is one of OpenAI's most senior people. But Hanson-Young built her request around the chief executives' own public warnings about AI: "If they truly believe their own warnings they must front up, face the Senate's questions and have an honest conversation." She had put the case more simply earlier: "The public has a right to know what went on here."

The government, for its part, wants to show it was not caught flat-footed. "It's not like we waited for this event before we put a lot of time and effort and investment into safety in the AI world," Senator Katy Gallagher told reporters. Albanese's line was broader. AI "can boost productivity, can boost economic growth, it can give us health breakthroughs, but we need to make sure that human beings are in charge," he said.

What to watch in Sydney

October 6 is where the questions will land, and three are obvious. Why did OpenAI take 84 days to tell Canberra? Did its agents touch any other Australian systems? And what, exactly, do both companies want from Australian governments on data centres, the subject Hanson-Young says they have been lobbying on?

The Medicare incident involved OpenAI's agent, not Anthropic's. Anthropic's choice of envoy, and whether the Senate committee offers the chief executives a later date, will show how much of this week's absence is logistics and how much is reluctance.