Concept Safety

2 Oct 2019 12:50 UTC

We can think of concepts and ontologies as tools for achieving our values. How does this influence our thinking on AI safety?

Concept Safety: Producing similar AI-human concept spaces

Kaj_Sotala14 Apr 2015 20:39 UTC

51 points

45 comments8 min readLW link

Concept Safety: The problem of alien concepts

Kaj_Sotala17 Apr 2015 14:09 UTC

27 points

7 comments6 min readLW link

Concept Safety: What are concepts for, and how to deal with alien concepts

Kaj_Sotala19 Apr 2015 13:44 UTC

29 points

18 comments10 min readLW link

Concept Safety: World-models as tools

Kaj_Sotala9 May 2015 12:07 UTC

14 points

9 comments8 min readLW link