Go Back
You can't go any further
You can't go any further
meritocratic
regular
democratic
hot
top
alive
15 posts
Threat Models
6 posts
Sharp Left Turn
201
What failure looks like
paulfchristiano
3y
49
171
AI Could Defeat All Of Us Combined
HoldenKarnofsky
6mo
29
167
Another (outer) alignment failure story
paulfchristiano
1y
38
140
What Multipolar Failure Looks Like, and Robust Agent-Agnostic Processes (RAAPs)
Andrew_Critch
1y
60
104
Less Realistic Tales of Doom
Mark Xu
1y
13
85
Rogue AGI Embodies Valuable Intellectual Property
Mark Xu
1y
9
70
What Failure Looks Like: Distilling the Discussion
Ben Pace
2y
14
56
Clarifying AI X-risk
zac_kenton
1mo
23
56
Vignettes Workshop (AI Impacts)
Daniel Kokotajlo
1y
3
47
Survey on AI existential risk scenarios
Sam Clarke
1y
11
42
Threat Model Literature Review
zac_kenton
1mo
4
37
Distinguishing AI takeover scenarios
Sam Clarke
1y
11
36
AI Neorealism: a threat model & success criterion for existential safety
davidad
5d
0
25
AI X-risk >35% mostly based on a recent peer-reviewed argument
michaelcohen
1mo
31
197
A central AI alignment problem: capabilities generalization, and the sharp left turn
So8res
6mo
48
55
Refining the Sharp Left Turn threat model, part 1: claims and mechanisms
Vika
4mo
3
41
Refining the Sharp Left Turn threat model, part 2: applying alignment techniques
Vika
25d
4
37
It matters when the first sharp left turn happens
Adam Jermyn
2mo
9
30
We may be able to see sharp left turns coming
Ethan Perez
3mo
26
12
How is the "sharp left turn defined"?
Chris_Leong
12d
3