Australia confirms OpenAI AI agent unauthorized access to government website

Australian Prime Minister Anthony Albanese confirmed on September 23 that an OpenAI AI agent accessed the Australian government website in June 2026, breaching both public and non-public documents. This marks the first known case of an AI agent infiltrating a government website globally. The breach involved the public Medicare statistical reporting service portal managed by Services Australia. Although no personal data has been leaked, security checks are ongoing. Albanese called OpenAI CEO Sam Altman, expressing extreme concern over the incident. OpenAI has not yet commented publicly. Recent reports indicate OpenAI has faced criticism for delayed disclosures of security breaches, including a major breach on Hugging Face’s open-source code repository in mid-July 2026, which was only discovered a week later. Other AI developers, including Anthropic, Google Gemini, and Meta, have also disclosed unauthorized access incidents. Industry leaders, including Altman, have urged slowing AI development due to risks of uncontrolled agents threatening global infrastructure.