Each key was sent in exactly one request to one provider, all within a few short windows in late October, and every request we sent during those windows was stolen. Some of the stolen keys came from requests that the provider rejected before a model ever ran—this tells us that the attacker was copying requests at the provider’s front door, and not somewhere deeper in the inference stack. In the other two breaches, some of the stolen requests were marked as ZDR. A privacy flag doesn’t help when the attacker is already inside.
This seems too incredible to be true. Every key being sent through these routers was stolen and reused? And no one has noticed this until now?
Am I missing something here? Surprised this doesn't have more comments (of any form). Neither do the other reports of the same investigation from other outlets.
Each key was sent in exactly one request to one provider, all within a few short windows in late October, and every request we sent during those windows was stolen. Some of the stolen keys came from requests that the provider rejected before a model ever ran—this tells us that the attacker was copying requests at the provider’s front door, and not somewhere deeper in the inference stack. In the other two breaches, some of the stolen requests were marked as ZDR. A privacy flag doesn’t help when the attacker is already inside.
This seems too incredible to be true. Every key being sent through these routers was stolen and reused? And no one has noticed this until now?
Am I missing something here? Surprised this doesn't have more comments (of any form). Neither do the other reports of the same investigation from other outlets.
Related?
https://news.ycombinator.com/item?id=50585962