AI ALIGNMENT FORUMTags
AF

Power Seeking (AI)

EditHistorySubscribe
Discussion (0)
Help improve this page
EditHistorySubscribe
Discussion (0)
Help improve this page
Power Seeking (AI)
Random Tag
Contributors
2Raymond Arnold

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)
1
16Instrumental convergence in single-agent systems
Edouard Harris, Simon Suo
1y
4
1
16Categorical-measure-theoretic approach to optimal policies tending to seek power
jacek
9mo
0
1
7POWERplay: An open-source toolchain to study AI power-seeking
Edouard Harris
1y
0
2
65Parametrically retargetable decision-makers tend to seek power
Alex Turner
7mo
3
1
29Eli's review of "Is power-seeking AI an existential risk?"
elifland
1y
0
1
27Power-seeking can be probable and predictive for trained agents
Victoria Krakovna, janos
7mo
20
1
17Generalizing the Power-Seeking Theorems
Alex Turner
3y
3
1
15[AN #170]: Analyzing the argument for risk from power-seeking AI
Rohin Shah
2y
0
1
10Power-seeking for successive choices
Adam Shimi
2y
9
1
61The Waluigi Effect (mega-post)
Cleo Nardo
7mo
24
0
23My Overview of the AI Alignment Landscape: Threat Models
Neel Nanda
2y
2
1
13Incentives from a causal perspective
Tom Everitt, James Fox, Ryan Carey, Matt MacDermott, Sebastian Benthall, Jonathan Richens
3mo
0
Add Posts