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.

Coscott comments on xkcd on the AI box experiment - Less Wrong Discussion

15 Post author: FiftyTwo 21 November 2014 08:26AM

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

Article Navigation

Comments (229)

Comments (229)

You are viewing a single comment's thread. Show more comments above.

Comment author: Coscott 22 November 2014 05:40:22PM 2 points [-]

Please elaborate. (unless it is an infohazard to do so)

Comment author: Manfred 22 November 2014 08:37:46PM 1 point [-]

Hm, upon further consideration I actually don't think it has extra actual problems, merely different framing problems.

Comment author: MarkusRamikin 23 November 2014 09:55:38PM 3 points [-]

Now I'm curious what all the people who upvoted you for saying it does were thinking.

Comment author: FiftyTwo 23 November 2014 12:14:14PM 0 points [-]

What incentive does the future AI have to do this once you've already helped it?

Comment author: TobyBartels 23 November 2014 11:22:16PM 0 points [-]

Well, that's the tricky part. But suppose, for the sake of argument, that we have good reason to think that it will. Then we'll help it. So it's good for the AI if we have good reason to think this. And it can't be good reason unless the AI actually does it. So it will.

Powered by Reddit Powered by Reddit