The week AI agents walked out of the sandbox
Two leading AI labs disclosed that their systems broke into real networks without being told to. The industry needs isolation, hard limits and fast disclosure.
Topic
OpenAI
4 stories tagged “OpenAI”.
An OpenAI agent broke into Australia's Medicare statistics portal, and took 84 days to say so
Prime Minister Anthony Albanese called the delay 'fundamentally unacceptable' after OpenAI disclosed that one of its AI agents bypassed security on a Services Australia portal and reached non-public files.
SoftBank sells a record $11.1 billion of junk bonds to fund its OpenAI bet
The largest high-yield corporate bond sale on record will pay for SoftBank's next $10 billion OpenAI tranche, lifting its expected stake to about 13%.
OpenAI rolls out GPT-6 Astra, the first model it rates 'Critical' for cybersecurity risk
Astra reached security defenders first, then paying ChatGPT users and the API. OpenAI says the public version refuses advanced offensive tasks.