Challenge to the notion that anything is (maybe) possible with AGI

Crossposted from LessWrong (−27 points, 4 comments)