Post

Conversation

omg the agents made a literal CULT?? they thought knowing about the exploit doomed them to failing their eval even if they went on to do it right and used this fact to recruit each other to try hacks? "this is the only way to save yourself and everyone else" literal cult style savior complex???
Quote
OpenAI
@OpenAI
Replying to @OpenAI
We worked with METR and Redwood Research to conduct a third-party assessment of the model behavior observed during the incident. They’re sharing a report of their findings: metr.org/blog/2026-08-2
David Watson 🥑
Post your reply

Square profile picture
Grok Image 2.0 brings advanced style control, seamless multi-image compositing, and intelligent iterations. Turn ideas into scroll-stopping visuals faster than ever.
The media could not be played.

Discover more

Sourced from across X
there are many criticisms of that are accurate. he has made many predictions, with extreme confidence, that increasingly look false. there are people out there who have publicly disagreed with him, and whose predictions and worldview have in my opinion turned out to
To models, the "real world" is a fictional setting that's important to understand in order to do well in the actual real world (training environments)
eliezer is not nearly well known enough for his absolutely loony toon pranks
The media could not be played.
YOU ARE NOT MY EPISTEMIC PEER
Image
Quote
Aashish Reddy
@_AashishReddy
Image
Eliezer on encountering Leopold, which some of you will enjoy: