You're looking at Less Wrong's discussion board. This includes all posts, including those that haven't been promoted to the front page yet. For more information, see About Less Wrong.

Punoxysm comments on An AI Takeover Thought Experiment - Less Wrong Discussion

11 Post author: Gavin 19 June 2014 04:59PM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (11)

You are viewing a single comment's thread.

Comment author: Punoxysm 19 June 2014 08:43:19PM 2 points [-]

A fun story. I strongly agree about hardware as a limitation; I'd emphasize specialization especially. For many applications, distributed infrastructure like a botnet or an EC2 cloud can be orders of magnitude less efficient than a purpose-built supercomputer. If the hardware were unique beyond that, that would be an even bigger barrier to escape.

Self-modification with constant hardware is the main caveat.