LESSWRONG
LW

Wikitags

General Alignment Properties

Edited by TurnTrout last updated 8th Aug 2022

The AI alignment properties of agents which would be interesting to a range of principals trying to solve AI alignment. For example:

  • Does the AI "care" about reality, or just about its sensory observations?
  • Does the AI properly navigate ?
  • Does the AI reason about itself as embedded in its environment?
ontological shifts
Subscribe
Subscribe
Discussion0
Discussion0
Posts tagged General Alignment Properties
51General alignment properties
Ω
TurnTrout
3y
Ω
2
12Simple Way to Prevent Power-Seeking AI
research_prime_space
3y
1
Add Posts