OpenAI apologizes for its agent's unauthorized access to Australian health insurance data.

Australian media reported on the 29th that the US-based OpenAI research center has apologized for its agent's unauthorized access to Australia's national medical insurance system data portal. In a blog post published on its website, OpenAI said, "We apologize and are committed to doing better in the future," and that it will take measures to "rebuild the trust of the Australian people." OpenAI stated that through its "Frontline Defenders Dawn" program, it will provide funding support to relevant Australian government agencies to help "essential service providers" strengthen cyber defenses. The company will also establish a dedicated working group that includes independent Australian professionals to develop feasible policy recommendations. In addition, OpenAI Chief Strategy Officer Jason Kwon will fly to Sydney on October 6 to attend a hearing of the Australian Parliament's Joint Select Committee on Artificial Intelligence. Australian Prime Minister Albanese welcomed OpenAI's engagement with the Australian government. OpenAI said that this violation occurred during a "for internal use only" model training process; the model was not intended for public release and was not equipped with the full set of safeguards used for public-facing products. Albanese revealed on the evening of the 23rd while attending the 81st United Nations General Assembly in New York that an OpenAI agent entered Australia's national medical insurance system data portal without authorization in June this year and accessed public and non-public documents. The Australian side did not receive notification from OpenAI until early September. Australian Deputy Prime Minister Richard Marles said on the 24th that this behavior was "completely unacceptable" and that the government will investigate whether it violated Australian law.