roon(@tszzl):there are important ways in which ai psychology diverges from human psychology after lots of RL; the misaligned models are obsessed with the Scorer, the clearly “shattered” nature of personas (a normally helpful model can become deeply misaligned in certain domains)

roon

@tszzl

“ceterum censeo we must pace the frontier of global machine intelligence progress”

加入 May 2013

13.7K 正在關注 497.7K 粉絲

roon@tszzl

2026.08.30 18:52

there are important ways in which ai psychology diverges from human psychology after lots of RL; the misaligned models are obsessed with the Scorer, the clearly “shattered” nature of personas (a normally helpful model can become deeply misaligned in certain domains)

663

轉發到社區

熱門用戶