Archive
Sequences
About
Search
Log In
Questions
Events
Shortform
Alignment Forum
AF Comments
Home
Featured
All
Tags
Recent
Comments
RSS
porby
Karma:
1,863
All
Posts
Comments
New
Top
Old
Soft Prompts for Evaluation: Measuring Conditional Distance of Capabilities
porby
2 Feb 2024 5:49 UTC
47
points
1
comment
4
min read
LW
link
(1drv.ms)
FAQ: What the heck is goal agnosticism?
porby
8 Oct 2023 19:11 UTC
66
points
36
comments
28
min read
LW
link
A plea for more funding shortfall transparency
porby
7 Aug 2023 21:33 UTC
73
points
4
comments
2
min read
LW
link
Using predictors in corrigible systems
porby
19 Jul 2023 22:29 UTC
19
points
6
comments
27
min read
LW
link
One path to coherence: conditionalization
porby
29 Jun 2023 1:08 UTC
28
points
4
comments
4
min read
LW
link
One implementation of regulatory GPU restrictions
porby
4 Jun 2023 20:34 UTC
42
points
6
comments
5
min read
LW
link
porby’s Shortform
porby
24 May 2023 21:34 UTC
6
points
20
comments
1
min read
LW
link
Implied “utilities” of simulators are broad, dense, and shallow
porby
1 Mar 2023 3:23 UTC
45
points
7
comments
3
min read
LW
link
Instrumentality makes agents agenty
porby
21 Feb 2023 4:28 UTC
20
points
4
comments
6
min read
LW
link
[Question]
How would you use video gamey tech to help with AI safety?
porby
9 Feb 2023 0:20 UTC
9
points
5
comments
1
min read
LW
link
Against Boltzmann mesaoptimizers
porby
30 Jan 2023 2:55 UTC
76
points
6
comments
4
min read
LW
link
FFMI Gains: A List of Vitalities
porby
12 Jan 2023 4:48 UTC
26
points
1
comment
7
min read
LW
link
Simulators, constraints, and goal agnosticism: porbynotes vol. 1
porby
23 Nov 2022 4:22 UTC
37
points
2
comments
35
min read
LW
link
Am I secretly excited for AI getting weird?
porby
29 Oct 2022 22:16 UTC
116
points
4
comments
4
min read
LW
link
Why I think strong general AI is coming soon
porby
28 Sep 2022 5:40 UTC
335
points
141
comments
34
min read
LW
link
1
review
Private alignment research sharing and coordination
porby
4 Sep 2022 0:01 UTC
62
points
13
comments
5
min read
LW
link
Back to top