\n
LAST UPDATE
\n |
\n 24 Sep\n <\/p>\n
HELEN MCENTEE SAID the hacking of an Australian government website by a rogue OpenAI model highlights the need for \u201cus to be working together\u201d, as well as the need to implement safeguards.<\/p>\n
The foreign affairs minister\u2019s comments follow Australian prime minister Anthony Albanese telling media that the rogue OpenAI model bypassed safeguards during training and hacked the government website, as he admonished the ChatGPT creator over an \u201cobviously unacceptable\u201d breach.<\/p>\n
The AI tool sought access to a health statistics portal in June and \u201cdidn\u2019t accept no for an answer\u201d, sidestepping restrictions to breach a section hosting private files, Albanese said.<\/p>\n
OpenAI did not raise the alarm with the Australian government until September, when it sent a message to a generic email inbox that is only checked once each day.<\/p>\n
Global worries about the power of advanced AI tools are mounting after a string of hacking incidents involving models from both OpenAI and rival developer Anthropic.<\/p>\n
\u201cToday, I spoke with the CEO of OpenAI, Sam Altman, to express Australia\u2019s extreme concern about this incident,\u201d Albanese told reporters in New York on Wednesday.<\/p>\n
\u201cI also expressed my disappointment that it took the company way too long to inform the government what had occurred,\u201d he said.<\/p>\n
\u201cIt took until 10 September before there was any notification at all \u2013 and the notification was an email sent to just the public mailbox.\u201d<\/p>\n
The AI tool accessed public and non-public files hosted on an old health statistics website.<\/p>\n
Albanese said there was \u201cno evidence\u201d that personal information had been accessed and that other government services had not been compromised.<\/p>\n
\u201cNonetheless, this situation is obviously unacceptable.\u201d<\/p>\n