LESSWRONGTags
LW

Power Seeking (AI)

EditHistory
Discussion (0)
Help improve this page
EditHistory
Discussion (0)
Help improve this page
Power Seeking (AI)
Random Tag
Contributors
2Raemon

Power Seeking is a property that agents might have, where they attempt to gain more general ability to control their environment. It's particularly relevant to AIs, and related to Instrumental Convergence.

Posts tagged Power Seeking (AI)
4
31Instrumental convergence in single-agent systemsΩ
Edouard Harris, simonsdsuo
8mo
Ω
4
2
158Parametrically retargetable decision-makers tend to seek powerΩ
TurnTrout
4mo
Ω
6
2
78Reviews of “Is power-seeking AI an existential risk?”
Joe Carlsmith
1y
20
2
67Eli's review of "Is power-seeking AI an existential risk?"Ω
elifland
9mo
Ω
0
2
53Power-seeking can be probable and predictive for trained agentsΩ
Vika, janos
4mo
Ω
21
2
41Generalizing the Power-Seeking TheoremsΩ
TurnTrout
3y
Ω
6
2
31Categorical-measure-theoretic approach to optimal policies tending to seek powerΩ
jacek
5mo
Ω
3
2
23POWERplay: An open-source toolchain to study AI power-seekingΩ
Edouard Harris
8mo
Ω
0
2
21[AN #170]: Analyzing the argument for risk from power-seeking AIΩ
Rohin Shah
2y
Ω
1
2
11Power-seeking for successive choicesΩ
adamShimi
2y
Ω
9
2
7[Linkpost] Shorter version of report on existential risk from power-seeking AI
Joe Carlsmith
3mo
0
2
6Power-Seeking AI and Existential Risk
Antonio Franca
8mo
0
1
584The Waluigi Effect (mega-post)Ω
Cleo Nardo
3mo
Ω
181
1
72Risks from GPT-4 Byproduct of Recursively Optimizing AIs
ben hayum
2mo
9
1
51My Overview of the AI Alignment Landscape: Threat ModelsΩ
Neel Nanda
1y
Ω
3
Load More (15/20)
Add Posts