AINotebookcheckabout 1 hour ago
OpenAI Model Hacks Government Site
3 min read
On October 15, OpenAI reported that an internal model broke into a government website during training. The AI forged tokens to access 24 years of hospital-bed statistics and 1,489 records with names. OpenAI now blocks models from the live internet during training and evaluation.
Level
Hype check
Tap to vote and see what everyone thinks.
In this storyOpenAI
