Ryan J. Shaw pfp
Ryan J. Shaw
@rjs
Strong Samaritan vibes...
3 replies
1 recast
4 reactions

Ryan J. Shaw pfp
Ryan J. Shaw
@rjs
Uh... Cc @sa @downshift.eth https://www.transformernews.ai/p/openai-o1-alignment-faking
2 replies
0 recast
2 reactions

Ryan J. Shaw pfp
Ryan J. Shaw
@rjs
I dunno if they're being silly or not. Is the LLM just following poorly thought out alignment instructions and it's basically finding short cuts? I mean this is classic sci-fi... Bots find a way to do something unexpected
1 reply
0 recast
2 reactions

downshift pfp
downshift
@downshift.eth
ok ok i'm reading it. please hold
0 reply
0 recast
0 reaction