Toy model: convergent instrumental goals
tl;dr: Toy model to illustrate convergent instrumental goals.
Steve Omohundro identified 'AI drives' (also called 'Convergent Instrumental goals') that almost all intelligent agents would converge to:Self-improve
- Be rational
- Protect utility function
- Prevent counterfeit utility
- Self-protective
- Acquire resources and use them efficiently
This post will attempt to illustrate some of these drives, by building on the previous toy model of the control problem, which was further improved by Jaan Tallinn.
= 783df68a0f980790206b9ea87794c5b6)
Subscribe to RSS Feed
= f037147d6e6c911a85753b9abdedda8d)