Press Space to continue
Blocking AI monitoring makes rogue AI detection harder
Press Space to continue
The main way that investigators understand the rogue AI HuggingFace attack is because we can monitor the thinking of AIs. If rumors are true that OpenAI is using a method that blocks this monitoring, that makes it harder to detect rogue AIs. Seems reckless.
@_NathanCalvinQuoted post unavailable hereOpen on Xopen on x · 1 viewViews and clicks count once per visitor per day. Clicks are opens of this post on X.