This website requires javascript to properly function. Consider activating javascript to get access to all site functionality.
LESSWRONG
"Why Not Just..."
LW
Login
"Why Not Just..."
A compendium of rants about alignment proposals, of varying charitability.
133
Deep Learning Systems Are Not Less Interpretable Than Logic/Probability/Etc
Ω
johnswentworth
1y
Ω
52
144
Godzilla Strategies
Ω
johnswentworth
1y
Ω
66
75
Rant on Problem Factorization for Alignment
Ω
johnswentworth
10mo
Ω
48
119
Interpretability/Tool-ness/Alignment/Corrigibility are not Composable
Ω
johnswentworth
10mo
Ω
12
157
How To Go From Interpretability To Alignment: Just Retarget The Search
Ω
johnswentworth
10mo
Ω
32
90
Oversight Misses 100% of Thoughts The AI Does Not Think
Ω
johnswentworth
10mo
Ω
49
71
Human Mimicry Mainly Works When We’re Already Close
Ω
johnswentworth
10mo
Ω
16
175
Worlds Where Iterative Design Fails
Ω
johnswentworth
10mo
Ω
27
149
Why Not Just... Build Weak AI Tools For AI Alignment Research?
Ω
johnswentworth
3mo
Ω
17
123
Why Not Just Outsource Alignment Research To An AI?
Ω
johnswentworth
3mo
Ω
45