This website requires javascript to properly function. Consider activating javascript to get access to all site functionality.
LESSWRONG
Tags
LW
Login
Refine
Edit
History
Discussion
(0)
Help improve this page
Edit
History
Discussion
(0)
Help improve this page
Refine
Random Tag
Contributors
3
NeuralBets
2
Ruby
Refine
is a conceptual research incubator hosted by
Conjecture
.
Posts tagged
Refine
Most Relevant
4
33
confusion about alignment requirements
Ω
Tamsin Leake
8mo
Ω
10
4
19
Refine's Second Blog Post Day
Ω
adamShimi
10mo
Ω
0
3
32
What if we approach AI safety like a technical engineering safety problem
Ω
zeshen
10mo
Ω
4
3
25
Benchmarking Proposals on Risk Scenarios
Ω
Paul Bricman
10mo
Ω
2
2
104
What I Learned Running Refine
Ω
adamShimi
7mo
Ω
5
2
48
PreDCA: vanessa kosoy's alignment protocol
Ω
Tamsin Leake
10mo
Ω
8
2
39
the Insulated Goal-Program idea
Ω
Tamsin Leake
10mo
Ω
4
2
30
Epistemic Artefacts of (conceptual) AI alignment research
Ω
Nora_Ammann
,
particlemania
10mo
Ω
1
2
30
Representational Tethers: Tying AI Latents To Human Ones
Ω
Paul Bricman
9mo
Ω
0
2
27
ordering capability thresholds
Ω
Tamsin Leake
9mo
Ω
0
2
27
goal-program bricks
Ω
Tamsin Leake
10mo
Ω
2
2
23
Refine Blogpost Day #3: The shortforms I did write
Ω
Alexander Gietelink Oldenziel
9mo
Ω
0
2
18
Refine's Third Blog Post Day/Week
Ω
adamShimi
9mo
Ω
0
1
137
Refine: An Incubator for Conceptual Alignment Research Bets
Ω
adamShimi
1y
Ω
13
1
87
How to Diversify Conceptual Alignment: the Model Behind Refine
Ω
adamShimi
1y
Ω
11