= 27d567bc2d48d9f40e9ccc20ea370b15)
David_Gerard comments on AI Risk and Opportunity: A Strategic Analysis - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (161)
Has anyone constructed even a vaguely plausible outline, let alone a definition, of what would constitute a "human-friendly intelligence", defined in terms other than effects you don't want it to have? As you note, humans aren't human-friendly intelligences, or we wouldn't have internal existential risk.
The CEV proposal seems to attempt to move the hard bit to technological magic (a superintelligence scanning human brains and working out a solution to human desires that is possible, is coherent and won't destroy us all) - this is saying "then a miracle occurs" in more words.
It's possible that particular humans might approximate human friendly intelligences.
Assuming it's not impossible, how would you know? What constitutes a human-friendly intelligence, in other than negative terms?
Er, that's how it is defined - at least by Yudkowsky. You want to argue definitions? Without even offering one of your own? How will that help?