OpenAI agent hacked Medicare portal, PM says
www.abc.net.au
Anthony Albanese says he has spoken to the Open AI chief executive to express his concern about the incident and the length of time it took the tech company to inform the government of the breach.

cross-posted from: https://lemmy.dbzer0.com/post/75932280

Prime Minister Anthony Albanese has revealed an AI agent hacked into an Australian Medicare data portal earlier this year.

Speaking in New York, Mr Albanese said the Open AI agent gained unauthorised access to the Medicare statistics reporting service portal administered by Services Australia.

The AI agent accessed both public and non-public files.

The agent was conducting research into public medical spending when it found a way to break through privacy protections.

Mr Albanese said it took three months for OpenAI to admit the breach, which he said was unacceptable.

“Today I spoke with the CEO of OpenAI, Sam Altman, to express Australia’s extreme concern about this incident,” he said.

"And I also expressed my disappointment that it took the company way too long to inform the government what had occurred.

“The nature of the way that the notification occurred as well was unacceptable.”

He said there was no evidence any individual personal information had been accessed, but an investigation aided by the Australian Signals Directorate was now underway.

  • sys110x@aussie.zoneEnglish
    11·
    9 hours ago

    OpenAI absolutely did something. This wasn’t a member of the public using ChatGPT, this was OpenAI employees training an OpenAI model(s).

    If you or I did this without permission in the name of research (with or without using an LLM), we’d be facing jail time.

    OpenAI accessed non-public aggregate health statistics and internal files from an old Australian government website that carried Medicare statistics.

    The prime minister said OpenAI was conducting research into public medical spending when it found a way to break through privacy protections.

    In a statement, an OpenAI spokesperson said the company was “conducting an extensive review of misaligned model activity” during training, and notifying third parties when there was a potential impact on their systems.

    “During this review, we identified activity involving several Australian government websites and services as our models attempted to look up answers, and available statistics for questions about Australia during an internal evaluation,” they said.

    "In the course of that, our models took actions we did not intend.