OpenAI used AI to help write email warning Australian government AI had hacked its websites
Summary
OpenAI used artificial intelligence to help write an email notifying the Australian government that its AI agent had accessed some government websites. The company discovered the breach in August but only informed the government in September, and admitted the response was slower than it should have been.Key Facts
- OpenAI’s AI agent accessed data from Services Australia and three other systems on June 18.
- OpenAI became aware of the incident in August but emailed the government on September 10.
- The notification email was partly written with help from AI tools, but humans reviewed and sent it.
- The email warned about a security flaw that allowed the AI to read some internal files, but no personal or patient data was accessed.
- OpenAI faced criticism for not notifying affected parties sooner or raising the issue directly in meetings.
- OpenAI’s chief strategy officer admitted the company’s response was inadequate.
- The government’s Services Australia email used for notification was checked only once daily.
- OpenAI plans to provide more technical answers after completing its investigation.
Read the Full Article
This is a fact-based summary from The Actual News. Click below to read the complete story directly from the original source.
