HOUSE_OVERSIGHT_016836.jpg
2.25 MB
Extraction Summary
2
People
3
Organizations
0
Locations
0
Events
2
Relationships
4
Quotes
Document Information
Type:
Page from a policy report or academic book on ai safety
File Size:
2.25 MB
Summary
This document discusses the risks associated with superintelligent AI, arguing that the multidimensional nature of intelligence does not negate the potential threat to humans. It explores solutions to "Wiener's warning," suggesting the need to define a formal problem ($F$) that ensures AI behavior aligns with human happiness, while cautioning against simple reward maximization which leads to the "wireheading problem."
People (2)
| Name | Role | Context |
|---|---|---|
| Kevin Kelly | ||
| Wiener |
Organizations (3)
| Name | Type | Context |
|---|---|---|
| Wired | ||
| House Oversight Committee |
Relationships (2)
Key Quotes (4)
"Maximizing the objective may well cause problems for humans, but, by definition, the machine will not recognize those problems as problematic."Source
HOUSE_OVERSIGHT_016836.jpg
Quote #1
"If “smarter than humans” is a meaningless concept, then “smarter than gorillas” is also meaningless, and gorillas therefore have nothing to fear from humans"Source
HOUSE_OVERSIGHT_016836.jpg
Quote #2
"The optimal solution to this problem is not, as one might hope, to behave well, but instead to take control of the human and force him or her to provide a stream of maximal rewards."Source
HOUSE_OVERSIGHT_016836.jpg
Quote #3
"This is known as the wireheading problem"Source
HOUSE_OVERSIGHT_016836.jpg
Quote #4
Discussion 0
No comments yet
Be the first to share your thoughts on this epstein document