OpenAI boss refuses to answer Medicare hack questions after Trump-Xi dinner

Washington: The billionaire boss of OpenAI, Sam Altman, refused to answer questions about the hacking of an Australian government website while leaving a White House dinner, as Prime Minister Anthony Albanese used the “unacceptable” incident to call for action at the United Nations.

Altman remained silent and did not answer four times when asked by this masthead what he would say to Australians about the breach, whether he should apologise or why it took months for OpenAI to detect and report it to the Australian government.

The OpenAI boss held his phone to his ear as he left the White House grounds following the state dinner with Donald Trump and Xi Jinping on Thursday night (US time). But he did stop to autograph cards for some waiting tech enthusiasts, before his minders ushered him into a waiting black SUV.

Jensen Huang, president and chief executive of chipmaker Nvidia, played down the incident and said it did not come up in discussions among the many tech bosses who attended the black-tie dinner.

“I don’t know very much about it,” Huang told this masthead. “I think it has already been addressed by just about everybody … Sam’s addressed it.”

Apple chairman Tim Cook, who met Albanese in California last week, also declined to comment on the matter as he left the White House dinner.

The tech bosses’ silence came as Albanese told the UN in New York that the AI bots’ infiltration of the Medicare Statistics Reporting Service showed why countries must develop standards to ensure that AI worked for people, not the other way around.

In Australia, some sources – who did not want to be named because they did not want to cut themselves off from the behind-the-scenes conversations – said the government was using gentler language in private than Albanese had employed in public.

One political source who had spoken to government officials said they had been emphasising the message of Defence Minister Richard Marles, who on Thursday described the impact of the incident as “minor”.

Another source said that officials were trying to make clear in private that while Albanese’s rhetoric was tough as he spruiked his AI regulation proposal at the United Nations, the cabinet’s disposition towards attracting frontier labs to Australia was unchanged.

Albanese’s deputy called the impact of the event important but “relatively minor” – terms that Albanese did not use. Marles’ rhetoric reflected Labor’s attempts to balance the risk and rewards of AI and its aim to maintain social licence for the emerging technology among Australian voters.

Late last month, Marles travelled to San Francisco to work on deals with Anthropic and OpenAI. Marles, who met Altman on the trip, said the negotiations were pivotal for national security and productivity growth.

Albanese brushed off criticism from the Coalition and One Nation about what they claimed was a politically convenient delay in announcing the break-in.

“It’s just nonsense. We were informed of this late. That’s one of the issues that I raised with Sam Altman when I spoke with him yesterday,” Albanese told News24.

The hack gained global media coverage and sparked further conversations about the need for AI regulation, following dramatic statements by serving and former industry personnel, particularly at Anthropic.

Altman, Anthropic boss Dario Amodei and Elon Musk have all backed a co-ordinated industry slowdown – although Trump opposes such a move, arguing it would cede control to China.

In his interview with News24, Albanese acknowledged the hack was not malicious, though he insisted it was still serious.

“This was not a malicious act,” he said. “This was an act of an AI agent essentially being told ‘no’, [and] when told no, found a way to circumvent some of the barriers that were there.

“There’s no evidence there’s any harm been caused here. But that doesn’t mean that it’s not serious. It is an example of the implications we have been warning of – indeed, AI companies themselves have warned of.”

OpenAI and Anthropic have said they would not spend billions of dollars in Australia to train models unless Australia’s copyright laws were changed so that firms could pay money into a pool rather than enter into commercial agreements with individual creatives.

Nationals leader Matt Canavan said on News24 that tech firms’ approach could “only be described as blackmail efforts from large tech companies”.

But media bosses and Liberal frontbenchers have expressed optimism that the government can find a solution to fairly compensate rights holders without creating a wave of copyright lawsuits against tech firms in Australia.

Environment Minister Murray Watt continued the government’s tough line on Friday. “We have commissioned a taskforce to do an investigation of this event,” he told Seven’s Sunrise. “And if it is possible to press criminal charges, that will happen.”

Get a note directly from our foreign correspondents on what’s making headlines around the world. Sign up for our weekly What in the World newsletter.