Go Back
Choose this branch
Choose this branch
meritocratic
regular
democratic
hot
top
alive
41 posts
Impact Regularization
Exercises / Problem-Sets
Fixed Point Theorems
84 posts
World Modeling
Anthropics
AIXI
Updateless Decision Theory
Sleeping Beauty Paradox
Cognitive Science
Extraterrestrial Life
Economics
Grabby Aliens
Infinity
83
Reframing Impact
TurnTrout
3y
15
77
Deducing Impact
TurnTrout
3y
26
75
World State is the Wrong Abstraction for Impact
TurnTrout
3y
19
71
Attainable Utility Theory: Why Things Matter
TurnTrout
3y
24
71
Best reasons for pessimism about impact of impact measures?
TurnTrout
3y
55
66
Value Impact
TurnTrout
3y
8
66
Towards a New Impact Measure
TurnTrout
4y
159
65
Attainable Utility Preservation: Empirical Results
TurnTrout
2y
8
65
Appendix: how a subagent could get powerful
Stuart_Armstrong
2y
17
57
Worrying about the Vase: Whitelisting
TurnTrout
4y
26
56
Attainable Utility Landscape: How The World Is Changed
TurnTrout
2y
7
55
Topological Fixed Point Exercises
Scott Garrabrant
4y
52
50
The Catastrophic Convergence Conjecture
TurnTrout
2y
15
49
The Gears of Impact
TurnTrout
3y
16
573
Where I agree and disagree with Eliezer
paulfchristiano
6mo
205
239
Without specific countermeasures, the easiest path to transformative AI likely leads to AI takeover
Ajeya Cotra
5mo
89
159
My research methodology
paulfchristiano
1y
36
159
Testing The Natural Abstraction Hypothesis: Project Intro
johnswentworth
1y
34
145
Fixing The Good Regulator Theorem
johnswentworth
1y
25
100
Selection Theorems: A Program For Understanding Agents
johnswentworth
1y
23
98
There is essentially one best-validated theory of cognition.
abramdemski
1y
34
97
Frequent arguments about alignment
John Schulman
1y
16
84
Less Threat-Dependent Bargaining Solutions?? (3/2)
Diffractor
4mo
7
83
The Goldbach conjecture is probably correct; so was Fermat's last theorem
Stuart_Armstrong
2y
27
77
Abstractions as Redundant Information
johnswentworth
10mo
7
75
«Boundaries», Part 3a: Defining boundaries as directed Markov blankets
Andrew_Critch
1mo
13
72
UDT can learn anthropic probabilities
cousin_it
4y
10
71
Chu are you?
Adele Lopez
1y
7