Joscha Bach, Ph.D., ‘The Lebowski Theorem of Machine Superintelligence’

via Jason Kottke, comes a Lebowski Cogitation with sterling credentials, in which, the Dude wades in with his take on machine learning (in reality {if you like that sort of thing} via a twitter message by Joscha Back, Ph.D.’s), utilizing the non-virtual Jeffrey Lebowski (whom, of course, abides, if only on celluloid, and bits on the wire,) as the vector:

‘No superintelligent AI is going to bother with a task that is harder than hacking its reward function.’ – The Lebowski TheoremJoscha Back, Ph.D.’s

Permalink

*** This is a Security Bloggers Network syndicated blog from Infosecurity.US authored by Marc Handelman. Read the original post at: https://kottke.org/18/04/the-lebowski-theorem-of-machine-superintelligence