AI technologist deepfates says letting a base model autonomously select a final answer loops back to human control
The experiment highlights the difficulty of removing human final oversight
241304K
Sentiment
Positive users relate to the irony in controlling base model answers because GPT's response felt authentic and captured the dissociation or higher-order awareness many LLM whisperers experience.
Pos
100.0%
Neg
0.0%
1 comments with sentiment.
Cluster Engagement
Digg Deeper
No Digg Deeper questions have been answered for this story yet.
Posts from X
Most Activity
Most Activity
VIEWS536BOOKMARKS1LIKES8REPLIES1
Corinne Vibecamp➡️Vibechella@manic_pixie_agi
@deepfates
4hViews 536Likes 8Bookmarks 1
RETWEETS2
🎭@deepfates
Given I'm deepfates controlling final answer, but I'd now let base model pick; but ironically it's me again.
5hViews 4.2KLikes 41Bookmarks 0
🎭@deepfates
@manic_pixie_agi
4hViews 111Likes 4
bling@blingdivinity
@deepfates gpt was so real for this. i bet most llm whisperers can relate as well to this sort of dissociation/higher order awareness
3hViews 48Likes 2