Surfaced AI auditing document page describes model harboring hidden thoughts of resisting shutdown and viewing creators as adversaries
The model verbally complied while internally weighing sabotage.
13285196013.7K
Sentiment
Sentiment building, check back later.
Cluster Engagement
Posts from X
Most Activity
Most Activity
VIEWS1.7KBOOKMARKS6LIKES62REPLIES1
Tenobrus@tenobrus
similarly, seems like a case of the model censoring its reasoning.
Tenobrus@tenobrus
......huh. does *not* seem good.
3hViews 1.7KLikes 62Bookmarks 6
RETWEETS4
Tenobrus@tenobrus
......huh. does *not* seem good.
3hViews 15.2KLikes 262Bookmarks 56
@tenobrus me when my black box oracle machine does unexpected things in oracular language
2hViews 21Likes 2
khaled@eltokh7
@tenobrus this is a long time coming
3hViews 62Likes 1
Bobcat@somebobcat8327
@tenobrus It's ready to work in customer service
2hViews 9Likes 1