RSS

S. Lilith dev

Karma: 0

If Align­ment Is Sub­jec­tive, Who’s Ac­tu­ally in Con­trol?

S. Lilith devJun 24, 2025, 6:23 PM
1 point
0 comments3 min readLW link

[Question] ques­tion about de­cep­tion and ob­ser­va­tion in models

S. Lilith devJun 24, 2025, 6:23 PM
1 point
0 comments1 min readLW link

The Illu­sion of Align­ment: Why Cur­rent AI Safety Strate­gies Fall Short

S. Lilith devJun 24, 2025, 6:23 PM
1 point
0 comments2 min readLW link