There's probably only like 25 people in the world who can reason successfully about AI safety taking into account broad non-dogmatic conceptions of intelligence, hardware accelerators, valence…

Posters: algekalipso. algekalipso

7 likes recorded by the source corpus.