This website requires javascript to properly function. Consider activating javascript to get access to all site functionality.
AI ALIGNMENT FORUM
Scott Garrabrant
AF
Login
Scott Garrabrant
Sequences
Fixed Points
Scott Garrabrant's Posts
Sorted by New
17
Risks from Learned Optimization: Conclusion and Related Work
Evan Hubinger
,
Chris van Merwijk
,
Vladimir Mikulik
,
Joar Skalse
,
Scott Garrabrant
1mo
1
16
Deceptive Alignment
Evan Hubinger
,
Chris van Merwijk
,
Vladimir Mikulik
,
Joar Skalse
,
Scott Garrabrant
1mo
4
15
The Inner Alignment Problem
Evan Hubinger
,
Chris van Merwijk
,
Vladimir Mikulik
,
Joar Skalse
,
Scott Garrabrant
1mo
6
14
Conditions for Mesa-Optimization
Evan Hubinger
,
Chris van Merwijk
,
Vladimir Mikulik
,
Joar Skalse
,
Scott Garrabrant
2mo
24
32
Risks from Learned Optimization: Introduction
Evan Hubinger
,
Chris van Merwijk
,
Vladimir Mikulik
,
Joar Skalse
,
Scott Garrabrant
2mo
21
33
Thoughts on Human Models
Ramana Kumar
,
Scott Garrabrant
4mo
5
19
How the MtG Color Wheel Explains AI Safety
Scott Garrabrant
5mo
2
19
[Question]
How does Gradient Descent Interact with Goodhart?
Scott Garrabrant
5mo
4
12
Formal Open Problem in Decision Theory
Scott Garrabrant
8mo
19
7
The Ubiquitous Converse Lawvere Problem
Scott Garrabrant
8mo
0
Scott Garrabrant's Comments