@andrew_n_carr You can follow this citation for the actual theory, much better than a superficial second hand account. https://t.co/LndMpQdCBP
1,717 followers
7,272 followers
@BogdanIonutCir2 Oh yeah it was definitely not about RLHF, sorry that it wasn't clear that it was humor. Still good to have spotted important failure modes that far in advance though. Here's the paper: https://t.co/3WVjEDB6MG
7,272 followers
@StefanFSchubert Oh yeah it was definitely not about RLHF, sorry that it wasn't clear that it was humor. Still good to have spotted important failure modes that far in advance though. Here's the paper: https://t.co/3WVjEDB6MG
22,043 followers
@Malky0010 @MilesCranmer The term was popularized by Ben Goertzel after being suggested by Shane Legg as the title for a collection of review articles back in 2007: https://t.co/xviij2VvwX